diff --git a/adapter_config.json b/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/adapter_model.bin b/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..9d959aeaa72be12c730b665a117398f76bf4c7d6 --- /dev/null +++ b/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7738c4edc7b66e957a9a9e7474a3be85403c8e4f0c0a506d2806acd4fed40d5 +size 125374989 diff --git a/checkpoint-100/README.md b/checkpoint-100/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-100/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-100/adapter_config.json b/checkpoint-100/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-100/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-100/adapter_model.bin b/checkpoint-100/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..46e4886ff41bf2bd5f267b6d2f2fef8eaba0f4b3 --- /dev/null +++ b/checkpoint-100/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7347e9c013e30092db854c3ac37c2ef4c96d75fcb7d34e5e89d1908990ab6ecd +size 125374989 diff --git a/checkpoint-100/adapter_model/README.md b/checkpoint-100/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-100/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-100/adapter_model/adapter_config.json b/checkpoint-100/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-100/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-100/adapter_model/adapter_model.bin b/checkpoint-100/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..46e4886ff41bf2bd5f267b6d2f2fef8eaba0f4b3 --- /dev/null +++ b/checkpoint-100/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7347e9c013e30092db854c3ac37c2ef4c96d75fcb7d34e5e89d1908990ab6ecd +size 125374989 diff --git a/checkpoint-100/optimizer.pt b/checkpoint-100/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..21a42d23545b0830ed969448c6378a0bdd5dd4c8 --- /dev/null +++ b/checkpoint-100/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:058a5ad19149195b58aa034208fbdd62061b7b65b5390ebb1e9511728f639a05 +size 250681597 diff --git a/checkpoint-100/rng_state_0.pth b/checkpoint-100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..1836ee6d5c49fe365eb74c1cff008f0795694993 --- /dev/null +++ b/checkpoint-100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f705fa4f8554ef13cd80d123eedf6db4d4f919bd202643f96ddc18c3dd814b7 +size 21687 diff --git a/checkpoint-100/rng_state_1.pth b/checkpoint-100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..c9d4f18f7da346c5227bf4d04f7f92217cad38b8 --- /dev/null +++ b/checkpoint-100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5f5ab19cf68a2202e4f485a33dd80f259f214096cb1aedc35f82497d399c23b +size 21687 diff --git a/checkpoint-100/rng_state_2.pth b/checkpoint-100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..f8ffbc401b7045c19d5478bdcc21673fc06cea73 --- /dev/null +++ b/checkpoint-100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d83e95ec95416e9f35064e113bfb9c4945355d2485e15fc46e5e1537a116797e +size 21687 diff --git a/checkpoint-100/rng_state_3.pth b/checkpoint-100/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..804d352c5d5450f85f124dc44aed3a6afbc9a7cc --- /dev/null +++ b/checkpoint-100/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fca50d7e35436d374671518fd1c8fb59611821f6ae0d33a22404297046ed918 +size 21687 diff --git a/checkpoint-100/rng_state_4.pth b/checkpoint-100/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..c48a03d5010613d5a9eed27e46e453161032dfad --- /dev/null +++ b/checkpoint-100/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9390d773fa037a30354bc7b4930c2aea8b97604f719eafdc6fb1fc02270476d +size 21687 diff --git a/checkpoint-100/rng_state_5.pth b/checkpoint-100/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..a6930b99b34727eed72f3b5e9f5e1a0631547ac9 --- /dev/null +++ b/checkpoint-100/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ee4b285615dbe11b31ff42bf5c128fcdb41ded83ece781479e3a09d0e6f5b92 +size 21687 diff --git a/checkpoint-100/rng_state_6.pth b/checkpoint-100/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..ab8ae0d35ad8baedb31bc24c4c421f254ac624f5 --- /dev/null +++ b/checkpoint-100/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc1fd611cb8a8f256e5eddd3ab6031506920951ecc59147c9e8e06d2c999828e +size 21687 diff --git a/checkpoint-100/rng_state_7.pth b/checkpoint-100/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..10d28e450cc2dc2288075cd4c131f33a5c6cf2be --- /dev/null +++ b/checkpoint-100/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a240cfccdea4f018cd47f4a7637dad6b0634bd559c84724ec04d368401214c4 +size 21687 diff --git a/checkpoint-100/scheduler.pt b/checkpoint-100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..e2f7d7ed3a99b673ada2198bbb024bd62956fe75 --- /dev/null +++ b/checkpoint-100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b49bfe742f6b22f71c53deba157d146b3b5a2f75c884c46630a1c049d5f07826 +size 627 diff --git a/checkpoint-100/trainer_state.json b/checkpoint-100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..22a9413008b53e42eb5b974d8dea2f60dacfe428 --- /dev/null +++ b/checkpoint-100/trainer_state.json @@ -0,0 +1,635 @@ +{ + "best_metric": 1.541548728942871, + "best_model_checkpoint": "./pippa-sharegpt-13b-qlora/checkpoint-100", + "epoch": 1.0554089709762533, + "eval_steps": 50, + "global_step": 100, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 2e-05, + "loss": 1.9036, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 4e-05, + "loss": 1.9209, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 6e-05, + "loss": 1.9161, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 8e-05, + "loss": 1.8722, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 0.0001, + "loss": 1.866, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 0.00012, + "loss": 1.872, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 0.00014, + "loss": 1.9138, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 0.00016, + "loss": 1.8785, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 0.00018, + "loss": 1.8013, + "step": 9 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002, + "loss": 1.7859, + "step": 10 + }, + { + "epoch": 0.12, + "learning_rate": 0.00019999332998034514, + "loss": 1.7805, + "step": 11 + }, + { + "epoch": 0.13, + "learning_rate": 0.00019997332081116373, + "loss": 1.7716, + "step": 12 + }, + { + "epoch": 0.14, + "learning_rate": 0.00019993997516168689, + "loss": 1.7453, + "step": 13 + }, + { + "epoch": 0.15, + "learning_rate": 0.00019989329748023725, + "loss": 1.7306, + "step": 14 + }, + { + "epoch": 0.16, + "learning_rate": 0.00019983329399363598, + "loss": 1.7535, + "step": 15 + }, + { + "epoch": 0.17, + "learning_rate": 0.0001997599727063717, + "loss": 1.7811, + "step": 16 + }, + { + "epoch": 0.18, + "learning_rate": 0.000199673343399533, + "loss": 1.7455, + "step": 17 + }, + { + "epoch": 0.19, + "learning_rate": 0.00019957341762950344, + "loss": 1.748, + "step": 18 + }, + { + "epoch": 0.2, + "learning_rate": 0.0001994602087264201, + "loss": 1.723, + "step": 19 + }, + { + "epoch": 0.21, + "learning_rate": 0.00019933373179239502, + "loss": 1.7539, + "step": 20 + }, + { + "epoch": 0.22, + "learning_rate": 0.000199194003699501, + "loss": 1.7293, + "step": 21 + }, + { + "epoch": 0.23, + "learning_rate": 0.0001990410430875205, + "loss": 1.6857, + "step": 22 + }, + { + "epoch": 0.24, + "learning_rate": 0.0001988748703614594, + "loss": 1.7209, + "step": 23 + }, + { + "epoch": 0.25, + "learning_rate": 0.00019869550768882455, + "loss": 1.7127, + "step": 24 + }, + { + "epoch": 0.26, + "learning_rate": 0.0001985029789966671, + "loss": 1.7024, + "step": 25 + }, + { + "epoch": 0.27, + "learning_rate": 0.0001982973099683902, + "loss": 1.6927, + "step": 26 + }, + { + "epoch": 0.28, + "learning_rate": 0.00019807852804032305, + "loss": 1.6584, + "step": 27 + }, + { + "epoch": 0.3, + "learning_rate": 0.0001978466623980609, + "loss": 1.7149, + "step": 28 + }, + { + "epoch": 0.31, + "learning_rate": 0.00019760174397257156, + "loss": 1.6815, + "step": 29 + }, + { + "epoch": 0.32, + "learning_rate": 0.0001973438054360693, + "loss": 1.7067, + "step": 30 + }, + { + "epoch": 0.33, + "learning_rate": 0.00019707288119765623, + "loss": 1.6799, + "step": 31 + }, + { + "epoch": 0.34, + "learning_rate": 0.00019678900739873226, + "loss": 1.7206, + "step": 32 + }, + { + "epoch": 0.35, + "learning_rate": 0.0001964922219081738, + "loss": 1.6458, + "step": 33 + }, + { + "epoch": 0.36, + "learning_rate": 0.00019618256431728194, + "loss": 1.6649, + "step": 34 + }, + { + "epoch": 0.37, + "learning_rate": 0.00019586007593450097, + "loss": 1.6684, + "step": 35 + }, + { + "epoch": 0.38, + "learning_rate": 0.000195524799779908, + "loss": 1.7411, + "step": 36 + }, + { + "epoch": 0.39, + "learning_rate": 0.00019517678057947384, + "loss": 1.6678, + "step": 37 + }, + { + "epoch": 0.4, + "learning_rate": 0.0001948160647590966, + "loss": 1.6738, + "step": 38 + }, + { + "epoch": 0.41, + "learning_rate": 0.00019444270043840852, + "loss": 1.6804, + "step": 39 + }, + { + "epoch": 0.42, + "learning_rate": 0.00019405673742435678, + "loss": 1.6728, + "step": 40 + }, + { + "epoch": 0.43, + "learning_rate": 0.00019365822720455916, + "loss": 1.6941, + "step": 41 + }, + { + "epoch": 0.44, + "learning_rate": 0.00019324722294043558, + "loss": 1.6861, + "step": 42 + }, + { + "epoch": 0.45, + "learning_rate": 0.00019282377946011652, + "loss": 1.6751, + "step": 43 + }, + { + "epoch": 0.46, + "learning_rate": 0.0001923879532511287, + "loss": 1.7175, + "step": 44 + }, + { + "epoch": 0.47, + "learning_rate": 0.00019193980245285966, + "loss": 1.6454, + "step": 45 + }, + { + "epoch": 0.49, + "learning_rate": 0.0001914793868488021, + "loss": 1.6934, + "step": 46 + }, + { + "epoch": 0.5, + "learning_rate": 0.0001910067678585786, + "loss": 1.6534, + "step": 47 + }, + { + "epoch": 0.51, + "learning_rate": 0.00019052200852974819, + "loss": 1.6797, + "step": 48 + }, + { + "epoch": 0.52, + "learning_rate": 0.00019002517352939598, + "loss": 1.6566, + "step": 49 + }, + { + "epoch": 0.53, + "learning_rate": 0.00018951632913550626, + "loss": 1.657, + "step": 50 + }, + { + "epoch": 0.53, + "eval_loss": 1.653587818145752, + "eval_runtime": 200.2473, + "eval_samples_per_second": 6.672, + "eval_steps_per_second": 0.21, + "step": 50 + }, + { + "epoch": 0.54, + "learning_rate": 0.0001889955432281212, + "loss": 1.6736, + "step": 51 + }, + { + "epoch": 0.55, + "learning_rate": 0.00018846288528028555, + "loss": 1.6696, + "step": 52 + }, + { + "epoch": 0.56, + "learning_rate": 0.00018791842634877898, + "loss": 1.6429, + "step": 53 + }, + { + "epoch": 0.57, + "learning_rate": 0.00018736223906463696, + "loss": 1.608, + "step": 54 + }, + { + "epoch": 0.58, + "learning_rate": 0.00018679439762346185, + "loss": 1.6245, + "step": 55 + }, + { + "epoch": 0.59, + "learning_rate": 0.00018621497777552507, + "loss": 1.6199, + "step": 56 + }, + { + "epoch": 0.6, + "learning_rate": 0.00018562405681566216, + "loss": 1.7028, + "step": 57 + }, + { + "epoch": 0.61, + "learning_rate": 0.00018502171357296144, + "loss": 1.6372, + "step": 58 + }, + { + "epoch": 0.62, + "learning_rate": 0.00018440802840024822, + "loss": 1.6198, + "step": 59 + }, + { + "epoch": 0.63, + "learning_rate": 0.00018378308316336584, + "loss": 1.6061, + "step": 60 + }, + { + "epoch": 0.64, + "learning_rate": 0.00018314696123025454, + "loss": 1.6051, + "step": 61 + }, + { + "epoch": 0.65, + "learning_rate": 0.00018249974745983023, + "loss": 1.6257, + "step": 62 + }, + { + "epoch": 0.66, + "learning_rate": 0.00018184152819066435, + "loss": 1.6295, + "step": 63 + }, + { + "epoch": 0.68, + "learning_rate": 0.00018117239122946615, + "loss": 1.6382, + "step": 64 + }, + { + "epoch": 0.69, + "learning_rate": 0.0001804924258393692, + "loss": 1.6032, + "step": 65 + }, + { + "epoch": 0.7, + "learning_rate": 0.000179801722728024, + "loss": 1.652, + "step": 66 + }, + { + "epoch": 0.71, + "learning_rate": 0.00017910037403549693, + "loss": 1.6431, + "step": 67 + }, + { + "epoch": 0.72, + "learning_rate": 0.00017838847332197938, + "loss": 1.6482, + "step": 68 + }, + { + "epoch": 0.73, + "learning_rate": 0.00017766611555530636, + "loss": 1.6276, + "step": 69 + }, + { + "epoch": 0.74, + "learning_rate": 0.00017693339709828792, + "loss": 1.611, + "step": 70 + }, + { + "epoch": 0.75, + "learning_rate": 0.00017619041569585418, + "loss": 1.6309, + "step": 71 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001754372704620164, + "loss": 1.6316, + "step": 72 + }, + { + "epoch": 0.77, + "learning_rate": 0.00017467406186664474, + "loss": 1.5779, + "step": 73 + }, + { + "epoch": 0.78, + "learning_rate": 0.00017390089172206592, + "loss": 1.6015, + "step": 74 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001731178631694811, + "loss": 1.6133, + "step": 75 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017232508066520702, + "loss": 1.6048, + "step": 76 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017152264996674136, + "loss": 1.5792, + "step": 77 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017071067811865476, + "loss": 1.5756, + "step": 78 + }, + { + "epoch": 0.83, + "learning_rate": 0.00016988927343831095, + "loss": 1.5497, + "step": 79 + }, + { + "epoch": 0.84, + "learning_rate": 0.00016905854550141716, + "loss": 1.5604, + "step": 80 + }, + { + "epoch": 0.85, + "learning_rate": 0.00016821860512740671, + "loss": 1.5722, + "step": 81 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016736956436465573, + "loss": 1.5943, + "step": 82 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016651153647553567, + "loss": 1.538, + "step": 83 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016564463592130428, + "loss": 1.5875, + "step": 84 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001647689783468362, + "loss": 1.5835, + "step": 85 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016388468056519612, + "loss": 1.5724, + "step": 86 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016299186054205577, + "loss": 1.573, + "step": 87 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016209063737995715, + "loss": 1.543, + "step": 88 + }, + { + "epoch": 0.94, + "learning_rate": 0.00016118113130242432, + "loss": 1.5438, + "step": 89 + }, + { + "epoch": 0.95, + "learning_rate": 0.00016026346363792567, + "loss": 1.5711, + "step": 90 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015933775680368822, + "loss": 1.5542, + "step": 91 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015840413428936767, + "loss": 1.5209, + "step": 92 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001574627206405744, + "loss": 1.569, + "step": 93 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001565136414422592, + "loss": 1.5657, + "step": 94 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015555702330196023, + "loss": 1.5248, + "step": 95 + }, + { + "epoch": 1.01, + "learning_rate": 0.00015459299383291345, + "loss": 1.5482, + "step": 96 + }, + { + "epoch": 1.02, + "learning_rate": 0.000153621681637029, + "loss": 1.5186, + "step": 97 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001526432162877356, + "loss": 1.5363, + "step": 98 + }, + { + "epoch": 1.04, + "learning_rate": 0.00015165772831269547, + "loss": 1.5221, + "step": 99 + }, + { + "epoch": 1.06, + "learning_rate": 0.00015066534917639195, + "loss": 1.5071, + "step": 100 + }, + { + "epoch": 1.06, + "eval_loss": 1.541548728942871, + "eval_runtime": 200.1636, + "eval_samples_per_second": 6.675, + "eval_steps_per_second": 0.21, + "step": 100 + } + ], + "logging_steps": 1, + "max_steps": 282, + "num_train_epochs": 3, + "save_steps": 50, + "total_flos": 1.029096231126696e+18, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-100/training_args.bin b/checkpoint-100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..41fe038f6bd77e41e63090845cfb2375d33e29ec --- /dev/null +++ b/checkpoint-100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8bdde5c34937c67d70adb2cea431eda05b8cf22eb5ebad425d871a634a8fc84 +size 4411 diff --git a/checkpoint-150/README.md b/checkpoint-150/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-150/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-150/adapter_config.json b/checkpoint-150/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-150/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-150/adapter_model.bin b/checkpoint-150/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..6731eb4ccf85b4d0983c9fc04cd2ba9513ef5474 --- /dev/null +++ b/checkpoint-150/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6282741cfdfa834dd0b8cd367070812ab0cc5a75ea9d41ef6dba922444c9625 +size 125374989 diff --git a/checkpoint-150/adapter_model/README.md b/checkpoint-150/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-150/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-150/adapter_model/adapter_config.json b/checkpoint-150/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-150/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-150/adapter_model/adapter_model.bin b/checkpoint-150/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..6731eb4ccf85b4d0983c9fc04cd2ba9513ef5474 --- /dev/null +++ b/checkpoint-150/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6282741cfdfa834dd0b8cd367070812ab0cc5a75ea9d41ef6dba922444c9625 +size 125374989 diff --git a/checkpoint-150/optimizer.pt b/checkpoint-150/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..b5d4a77e7f32d767bea13c48cd4aad4de4884615 --- /dev/null +++ b/checkpoint-150/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce6dcd2983bc5a32bd302a66647bda032b0572b986f4904bd4543d6d2d43b067 +size 250681597 diff --git a/checkpoint-150/rng_state_0.pth b/checkpoint-150/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..b615a1b2a9105231f74e405e039e79f1da013a9c --- /dev/null +++ b/checkpoint-150/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d209f7a1508a25c574e4c3d69a27ca41cba4cc2058612304668b6f5e97ab3d2 +size 21687 diff --git a/checkpoint-150/rng_state_1.pth b/checkpoint-150/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..84b6dec690497f282af065683e7aad53569cb100 --- /dev/null +++ b/checkpoint-150/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c15ad3c1a375d56daf983bf7705ee84d266d43306e2844bcbb660fc1cc0ca51 +size 21687 diff --git a/checkpoint-150/rng_state_2.pth b/checkpoint-150/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..d41a838f4628435ab631214f04e916af867ac827 --- /dev/null +++ b/checkpoint-150/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e6020aae6bf4cd66952300aaec4db5e3bc2938e89d85342eb6776cbb021072d +size 21687 diff --git a/checkpoint-150/rng_state_3.pth b/checkpoint-150/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..ac653f6802d417e71bc0a7017a66342b1679abc5 --- /dev/null +++ b/checkpoint-150/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40fc11f613f0cc52561363af8adcdefc232c2ece7688652a0073fa4f9ab6e6c0 +size 21687 diff --git a/checkpoint-150/rng_state_4.pth b/checkpoint-150/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..07a2babeb2f0470973ba34c61197cdd2af5d0a56 --- /dev/null +++ b/checkpoint-150/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a03a4a7816676c3ba781d35b7283b034c62c299d4d29b637df3f2c31a25a2fe +size 21687 diff --git a/checkpoint-150/rng_state_5.pth b/checkpoint-150/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..97267cf8e280180f6a45c9cb76db652494fe60ce --- /dev/null +++ b/checkpoint-150/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37063ab98ea8f2f61c649feaee1aca9a1cddb2dc025aa359bbb47a7ed22e7955 +size 21687 diff --git a/checkpoint-150/rng_state_6.pth b/checkpoint-150/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..6b89c865eee865c70a2d3678e7f36d4da6643580 --- /dev/null +++ b/checkpoint-150/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6277e18656ac27e03e4b3090eac4beefadf98706b399a7a20e75b54b1c2696e +size 21687 diff --git a/checkpoint-150/rng_state_7.pth b/checkpoint-150/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..9ca0d89357be14cbe5a60262bce66d54997dd8c3 --- /dev/null +++ b/checkpoint-150/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da8748808b2d5076a715a9ba94e767ae7a62e26de5bb5b4cb33229d479203300 +size 21687 diff --git a/checkpoint-150/scheduler.pt b/checkpoint-150/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..ae0997adfd6e7bd608f545973d6290a877343f00 --- /dev/null +++ b/checkpoint-150/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db732cfc752aa95d11e53afbebc7fa041457f5539315097963802e4d8585d404 +size 627 diff --git a/checkpoint-150/trainer_state.json b/checkpoint-150/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..46c2217d8c80f7173e033ca52a106eb5beaf5578 --- /dev/null +++ b/checkpoint-150/trainer_state.json @@ -0,0 +1,943 @@ +{ + "best_metric": 1.4968100786209106, + "best_model_checkpoint": "./pippa-sharegpt-13b-qlora/checkpoint-150", + "epoch": 1.58311345646438, + "eval_steps": 50, + "global_step": 150, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 2e-05, + "loss": 1.9036, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 4e-05, + "loss": 1.9209, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 6e-05, + "loss": 1.9161, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 8e-05, + "loss": 1.8722, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 0.0001, + "loss": 1.866, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 0.00012, + "loss": 1.872, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 0.00014, + "loss": 1.9138, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 0.00016, + "loss": 1.8785, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 0.00018, + "loss": 1.8013, + "step": 9 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002, + "loss": 1.7859, + "step": 10 + }, + { + "epoch": 0.12, + "learning_rate": 0.00019999332998034514, + "loss": 1.7805, + "step": 11 + }, + { + "epoch": 0.13, + "learning_rate": 0.00019997332081116373, + "loss": 1.7716, + "step": 12 + }, + { + "epoch": 0.14, + "learning_rate": 0.00019993997516168689, + "loss": 1.7453, + "step": 13 + }, + { + "epoch": 0.15, + "learning_rate": 0.00019989329748023725, + "loss": 1.7306, + "step": 14 + }, + { + "epoch": 0.16, + "learning_rate": 0.00019983329399363598, + "loss": 1.7535, + "step": 15 + }, + { + "epoch": 0.17, + "learning_rate": 0.0001997599727063717, + "loss": 1.7811, + "step": 16 + }, + { + "epoch": 0.18, + "learning_rate": 0.000199673343399533, + "loss": 1.7455, + "step": 17 + }, + { + "epoch": 0.19, + "learning_rate": 0.00019957341762950344, + "loss": 1.748, + "step": 18 + }, + { + "epoch": 0.2, + "learning_rate": 0.0001994602087264201, + "loss": 1.723, + "step": 19 + }, + { + "epoch": 0.21, + "learning_rate": 0.00019933373179239502, + "loss": 1.7539, + "step": 20 + }, + { + "epoch": 0.22, + "learning_rate": 0.000199194003699501, + "loss": 1.7293, + "step": 21 + }, + { + "epoch": 0.23, + "learning_rate": 0.0001990410430875205, + "loss": 1.6857, + "step": 22 + }, + { + "epoch": 0.24, + "learning_rate": 0.0001988748703614594, + "loss": 1.7209, + "step": 23 + }, + { + "epoch": 0.25, + "learning_rate": 0.00019869550768882455, + "loss": 1.7127, + "step": 24 + }, + { + "epoch": 0.26, + "learning_rate": 0.0001985029789966671, + "loss": 1.7024, + "step": 25 + }, + { + "epoch": 0.27, + "learning_rate": 0.0001982973099683902, + "loss": 1.6927, + "step": 26 + }, + { + "epoch": 0.28, + "learning_rate": 0.00019807852804032305, + "loss": 1.6584, + "step": 27 + }, + { + "epoch": 0.3, + "learning_rate": 0.0001978466623980609, + "loss": 1.7149, + "step": 28 + }, + { + "epoch": 0.31, + "learning_rate": 0.00019760174397257156, + "loss": 1.6815, + "step": 29 + }, + { + "epoch": 0.32, + "learning_rate": 0.0001973438054360693, + "loss": 1.7067, + "step": 30 + }, + { + "epoch": 0.33, + "learning_rate": 0.00019707288119765623, + "loss": 1.6799, + "step": 31 + }, + { + "epoch": 0.34, + "learning_rate": 0.00019678900739873226, + "loss": 1.7206, + "step": 32 + }, + { + "epoch": 0.35, + "learning_rate": 0.0001964922219081738, + "loss": 1.6458, + "step": 33 + }, + { + "epoch": 0.36, + "learning_rate": 0.00019618256431728194, + "loss": 1.6649, + "step": 34 + }, + { + "epoch": 0.37, + "learning_rate": 0.00019586007593450097, + "loss": 1.6684, + "step": 35 + }, + { + "epoch": 0.38, + "learning_rate": 0.000195524799779908, + "loss": 1.7411, + "step": 36 + }, + { + "epoch": 0.39, + "learning_rate": 0.00019517678057947384, + "loss": 1.6678, + "step": 37 + }, + { + "epoch": 0.4, + "learning_rate": 0.0001948160647590966, + "loss": 1.6738, + "step": 38 + }, + { + "epoch": 0.41, + "learning_rate": 0.00019444270043840852, + "loss": 1.6804, + "step": 39 + }, + { + "epoch": 0.42, + "learning_rate": 0.00019405673742435678, + "loss": 1.6728, + "step": 40 + }, + { + "epoch": 0.43, + "learning_rate": 0.00019365822720455916, + "loss": 1.6941, + "step": 41 + }, + { + "epoch": 0.44, + "learning_rate": 0.00019324722294043558, + "loss": 1.6861, + "step": 42 + }, + { + "epoch": 0.45, + "learning_rate": 0.00019282377946011652, + "loss": 1.6751, + "step": 43 + }, + { + "epoch": 0.46, + "learning_rate": 0.0001923879532511287, + "loss": 1.7175, + "step": 44 + }, + { + "epoch": 0.47, + "learning_rate": 0.00019193980245285966, + "loss": 1.6454, + "step": 45 + }, + { + "epoch": 0.49, + "learning_rate": 0.0001914793868488021, + "loss": 1.6934, + "step": 46 + }, + { + "epoch": 0.5, + "learning_rate": 0.0001910067678585786, + "loss": 1.6534, + "step": 47 + }, + { + "epoch": 0.51, + "learning_rate": 0.00019052200852974819, + "loss": 1.6797, + "step": 48 + }, + { + "epoch": 0.52, + "learning_rate": 0.00019002517352939598, + "loss": 1.6566, + "step": 49 + }, + { + "epoch": 0.53, + "learning_rate": 0.00018951632913550626, + "loss": 1.657, + "step": 50 + }, + { + "epoch": 0.53, + "eval_loss": 1.653587818145752, + "eval_runtime": 200.2473, + "eval_samples_per_second": 6.672, + "eval_steps_per_second": 0.21, + "step": 50 + }, + { + "epoch": 0.54, + "learning_rate": 0.0001889955432281212, + "loss": 1.6736, + "step": 51 + }, + { + "epoch": 0.55, + "learning_rate": 0.00018846288528028555, + "loss": 1.6696, + "step": 52 + }, + { + "epoch": 0.56, + "learning_rate": 0.00018791842634877898, + "loss": 1.6429, + "step": 53 + }, + { + "epoch": 0.57, + "learning_rate": 0.00018736223906463696, + "loss": 1.608, + "step": 54 + }, + { + "epoch": 0.58, + "learning_rate": 0.00018679439762346185, + "loss": 1.6245, + "step": 55 + }, + { + "epoch": 0.59, + "learning_rate": 0.00018621497777552507, + "loss": 1.6199, + "step": 56 + }, + { + "epoch": 0.6, + "learning_rate": 0.00018562405681566216, + "loss": 1.7028, + "step": 57 + }, + { + "epoch": 0.61, + "learning_rate": 0.00018502171357296144, + "loss": 1.6372, + "step": 58 + }, + { + "epoch": 0.62, + "learning_rate": 0.00018440802840024822, + "loss": 1.6198, + "step": 59 + }, + { + "epoch": 0.63, + "learning_rate": 0.00018378308316336584, + "loss": 1.6061, + "step": 60 + }, + { + "epoch": 0.64, + "learning_rate": 0.00018314696123025454, + "loss": 1.6051, + "step": 61 + }, + { + "epoch": 0.65, + "learning_rate": 0.00018249974745983023, + "loss": 1.6257, + "step": 62 + }, + { + "epoch": 0.66, + "learning_rate": 0.00018184152819066435, + "loss": 1.6295, + "step": 63 + }, + { + "epoch": 0.68, + "learning_rate": 0.00018117239122946615, + "loss": 1.6382, + "step": 64 + }, + { + "epoch": 0.69, + "learning_rate": 0.0001804924258393692, + "loss": 1.6032, + "step": 65 + }, + { + "epoch": 0.7, + "learning_rate": 0.000179801722728024, + "loss": 1.652, + "step": 66 + }, + { + "epoch": 0.71, + "learning_rate": 0.00017910037403549693, + "loss": 1.6431, + "step": 67 + }, + { + "epoch": 0.72, + "learning_rate": 0.00017838847332197938, + "loss": 1.6482, + "step": 68 + }, + { + "epoch": 0.73, + "learning_rate": 0.00017766611555530636, + "loss": 1.6276, + "step": 69 + }, + { + "epoch": 0.74, + "learning_rate": 0.00017693339709828792, + "loss": 1.611, + "step": 70 + }, + { + "epoch": 0.75, + "learning_rate": 0.00017619041569585418, + "loss": 1.6309, + "step": 71 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001754372704620164, + "loss": 1.6316, + "step": 72 + }, + { + "epoch": 0.77, + "learning_rate": 0.00017467406186664474, + "loss": 1.5779, + "step": 73 + }, + { + "epoch": 0.78, + "learning_rate": 0.00017390089172206592, + "loss": 1.6015, + "step": 74 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001731178631694811, + "loss": 1.6133, + "step": 75 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017232508066520702, + "loss": 1.6048, + "step": 76 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017152264996674136, + "loss": 1.5792, + "step": 77 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017071067811865476, + "loss": 1.5756, + "step": 78 + }, + { + "epoch": 0.83, + "learning_rate": 0.00016988927343831095, + "loss": 1.5497, + "step": 79 + }, + { + "epoch": 0.84, + "learning_rate": 0.00016905854550141716, + "loss": 1.5604, + "step": 80 + }, + { + "epoch": 0.85, + "learning_rate": 0.00016821860512740671, + "loss": 1.5722, + "step": 81 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016736956436465573, + "loss": 1.5943, + "step": 82 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016651153647553567, + "loss": 1.538, + "step": 83 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016564463592130428, + "loss": 1.5875, + "step": 84 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001647689783468362, + "loss": 1.5835, + "step": 85 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016388468056519612, + "loss": 1.5724, + "step": 86 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016299186054205577, + "loss": 1.573, + "step": 87 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016209063737995715, + "loss": 1.543, + "step": 88 + }, + { + "epoch": 0.94, + "learning_rate": 0.00016118113130242432, + "loss": 1.5438, + "step": 89 + }, + { + "epoch": 0.95, + "learning_rate": 0.00016026346363792567, + "loss": 1.5711, + "step": 90 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015933775680368822, + "loss": 1.5542, + "step": 91 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015840413428936767, + "loss": 1.5209, + "step": 92 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001574627206405744, + "loss": 1.569, + "step": 93 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001565136414422592, + "loss": 1.5657, + "step": 94 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015555702330196023, + "loss": 1.5248, + "step": 95 + }, + { + "epoch": 1.01, + "learning_rate": 0.00015459299383291345, + "loss": 1.5482, + "step": 96 + }, + { + "epoch": 1.02, + "learning_rate": 0.000153621681637029, + "loss": 1.5186, + "step": 97 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001526432162877356, + "loss": 1.5363, + "step": 98 + }, + { + "epoch": 1.04, + "learning_rate": 0.00015165772831269547, + "loss": 1.5221, + "step": 99 + }, + { + "epoch": 1.06, + "learning_rate": 0.00015066534917639195, + "loss": 1.5071, + "step": 100 + }, + { + "epoch": 1.06, + "eval_loss": 1.541548728942871, + "eval_runtime": 200.1636, + "eval_samples_per_second": 6.675, + "eval_steps_per_second": 0.21, + "step": 100 + }, + { + "epoch": 1.07, + "learning_rate": 0.00014966621126259183, + "loss": 1.5179, + "step": 101 + }, + { + "epoch": 1.08, + "learning_rate": 0.00014866044785668563, + "loss": 1.4923, + "step": 102 + }, + { + "epoch": 1.09, + "learning_rate": 0.00014764819312790707, + "loss": 1.5733, + "step": 103 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001466295821114348, + "loss": 1.4586, + "step": 104 + }, + { + "epoch": 1.11, + "learning_rate": 0.00014560475069037894, + "loss": 1.5379, + "step": 105 + }, + { + "epoch": 1.12, + "learning_rate": 0.00014457383557765386, + "loss": 1.5059, + "step": 106 + }, + { + "epoch": 1.13, + "learning_rate": 0.00014353697429774084, + "loss": 1.512, + "step": 107 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001424943051683422, + "loss": 1.4908, + "step": 108 + }, + { + "epoch": 1.15, + "learning_rate": 0.0001414459672819297, + "loss": 1.5067, + "step": 109 + }, + { + "epoch": 1.16, + "learning_rate": 0.00014039210048718949, + "loss": 1.4908, + "step": 110 + }, + { + "epoch": 1.17, + "learning_rate": 0.00013933284537036625, + "loss": 1.5299, + "step": 111 + }, + { + "epoch": 1.18, + "learning_rate": 0.000138268343236509, + "loss": 1.4836, + "step": 112 + }, + { + "epoch": 1.19, + "learning_rate": 0.00013719873609062077, + "loss": 1.5285, + "step": 113 + }, + { + "epoch": 1.2, + "learning_rate": 0.00013612416661871533, + "loss": 1.5493, + "step": 114 + }, + { + "epoch": 1.21, + "learning_rate": 0.0001350447781687826, + "loss": 1.4915, + "step": 115 + }, + { + "epoch": 1.22, + "learning_rate": 0.00013396071473166613, + "loss": 1.5139, + "step": 116 + }, + { + "epoch": 1.23, + "learning_rate": 0.00013287212092185464, + "loss": 1.4589, + "step": 117 + }, + { + "epoch": 1.25, + "learning_rate": 0.00013177914195819016, + "loss": 1.4869, + "step": 118 + }, + { + "epoch": 1.26, + "learning_rate": 0.00013068192364449618, + "loss": 1.5081, + "step": 119 + }, + { + "epoch": 1.27, + "learning_rate": 0.00012958061235012706, + "loss": 1.5172, + "step": 120 + }, + { + "epoch": 1.28, + "learning_rate": 0.0001284753549904423, + "loss": 1.5195, + "step": 121 + }, + { + "epoch": 1.29, + "learning_rate": 0.0001273662990072083, + "loss": 1.4635, + "step": 122 + }, + { + "epoch": 1.3, + "learning_rate": 0.00012625359234892907, + "loss": 1.5036, + "step": 123 + }, + { + "epoch": 1.31, + "learning_rate": 0.0001251373834511103, + "loss": 1.5203, + "step": 124 + }, + { + "epoch": 1.32, + "learning_rate": 0.00012401782121645766, + "loss": 1.5047, + "step": 125 + }, + { + "epoch": 1.33, + "learning_rate": 0.0001228950549950134, + "loss": 1.5047, + "step": 126 + }, + { + "epoch": 1.34, + "learning_rate": 0.00012176923456423284, + "loss": 1.4715, + "step": 127 + }, + { + "epoch": 1.35, + "learning_rate": 0.00012064051010900397, + "loss": 1.4884, + "step": 128 + }, + { + "epoch": 1.36, + "learning_rate": 0.00011950903220161285, + "loss": 1.4947, + "step": 129 + }, + { + "epoch": 1.37, + "learning_rate": 0.00011837495178165706, + "loss": 1.4665, + "step": 130 + }, + { + "epoch": 1.38, + "learning_rate": 0.00011723842013591044, + "loss": 1.4888, + "step": 131 + }, + { + "epoch": 1.39, + "learning_rate": 0.00011609958887814129, + "loss": 1.52, + "step": 132 + }, + { + "epoch": 1.4, + "learning_rate": 0.00011495860992888712, + "loss": 1.4305, + "step": 133 + }, + { + "epoch": 1.41, + "learning_rate": 0.00011381563549518823, + "loss": 1.4865, + "step": 134 + }, + { + "epoch": 1.42, + "learning_rate": 0.00011267081805028339, + "loss": 1.5155, + "step": 135 + }, + { + "epoch": 1.44, + "learning_rate": 0.00011152431031326978, + "loss": 1.4952, + "step": 136 + }, + { + "epoch": 1.45, + "learning_rate": 0.00011037626522873019, + "loss": 1.4765, + "step": 137 + }, + { + "epoch": 1.46, + "learning_rate": 0.00010922683594633021, + "loss": 1.4729, + "step": 138 + }, + { + "epoch": 1.47, + "learning_rate": 0.00010807617580038796, + "loss": 1.481, + "step": 139 + }, + { + "epoch": 1.48, + "learning_rate": 0.00010692443828941918, + "loss": 1.4877, + "step": 140 + }, + { + "epoch": 1.49, + "learning_rate": 0.00010577177705566061, + "loss": 1.4884, + "step": 141 + }, + { + "epoch": 1.5, + "learning_rate": 0.00010461834586457398, + "loss": 1.4786, + "step": 142 + }, + { + "epoch": 1.51, + "learning_rate": 0.00010346429858433352, + "loss": 1.4585, + "step": 143 + }, + { + "epoch": 1.52, + "learning_rate": 0.00010230978916530012, + "loss": 1.4867, + "step": 144 + }, + { + "epoch": 1.53, + "learning_rate": 0.00010115497161948409, + "loss": 1.487, + "step": 145 + }, + { + "epoch": 1.54, + "learning_rate": 0.0001, + "loss": 1.4693, + "step": 146 + }, + { + "epoch": 1.55, + "learning_rate": 9.884502838051595e-05, + "loss": 1.4391, + "step": 147 + }, + { + "epoch": 1.56, + "learning_rate": 9.76902108346999e-05, + "loss": 1.493, + "step": 148 + }, + { + "epoch": 1.57, + "learning_rate": 9.653570141566653e-05, + "loss": 1.5152, + "step": 149 + }, + { + "epoch": 1.58, + "learning_rate": 9.538165413542607e-05, + "loss": 1.4766, + "step": 150 + }, + { + "epoch": 1.58, + "eval_loss": 1.4968100786209106, + "eval_runtime": 199.9343, + "eval_samples_per_second": 6.682, + "eval_steps_per_second": 0.21, + "step": 150 + } + ], + "logging_steps": 1, + "max_steps": 282, + "num_train_epochs": 3, + "save_steps": 50, + "total_flos": 1.546341313689944e+18, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-150/training_args.bin b/checkpoint-150/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..41fe038f6bd77e41e63090845cfb2375d33e29ec --- /dev/null +++ b/checkpoint-150/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8bdde5c34937c67d70adb2cea431eda05b8cf22eb5ebad425d871a634a8fc84 +size 4411 diff --git a/checkpoint-200/README.md b/checkpoint-200/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-200/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-200/adapter_config.json b/checkpoint-200/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-200/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-200/adapter_model.bin b/checkpoint-200/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..2ae5bf43ea2a6b6afa54fe5cd2926d77794f8239 --- /dev/null +++ b/checkpoint-200/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42245ccd0d6a53485c7df59d7bda13e2c242bfa3855a784c1e3edba0d804af5e +size 125374989 diff --git a/checkpoint-200/adapter_model/README.md b/checkpoint-200/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-200/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-200/adapter_model/adapter_config.json b/checkpoint-200/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-200/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-200/adapter_model/adapter_model.bin b/checkpoint-200/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..2ae5bf43ea2a6b6afa54fe5cd2926d77794f8239 --- /dev/null +++ b/checkpoint-200/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42245ccd0d6a53485c7df59d7bda13e2c242bfa3855a784c1e3edba0d804af5e +size 125374989 diff --git a/checkpoint-200/optimizer.pt b/checkpoint-200/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..c61cdd3b25cf589603079a25fa7ff6b92cc880cf --- /dev/null +++ b/checkpoint-200/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddd52d3cb770a9644ca58a2924034ca29068888aa68d37df9d534df5a615d4bf +size 250681597 diff --git a/checkpoint-200/rng_state_0.pth b/checkpoint-200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..d146618edcdb8dd87b2115c8a32e07435f2e7482 --- /dev/null +++ b/checkpoint-200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a75a658d44fe38598390ec155c1a51d8fe99b3431b0a3b96be8d88bb3f3cb1b +size 21687 diff --git a/checkpoint-200/rng_state_1.pth b/checkpoint-200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..a272fb290779b711844017ae571a1f6f1e04fb03 --- /dev/null +++ b/checkpoint-200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:562b30f364e08187db5d3919ecd53271f4d3095698fba5dc692c4ed9cabe8b03 +size 21687 diff --git a/checkpoint-200/rng_state_2.pth b/checkpoint-200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..fdbf2647c882d798983f3d64d0bcc150117188f1 --- /dev/null +++ b/checkpoint-200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7075cf5206fca95776b844726640b7c2f6f7b89d5fbb4201cc739e018f95e8cb +size 21687 diff --git a/checkpoint-200/rng_state_3.pth b/checkpoint-200/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..06c23996cc1ca2e489979398d52084975c63d195 --- /dev/null +++ b/checkpoint-200/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0907d6758e3713a38066511dc86bd53d6a9524ea0a1bb2032c42fa2f66498ddb +size 21687 diff --git a/checkpoint-200/rng_state_4.pth b/checkpoint-200/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..bd3f85713cc7bc6bbe73348251cca6b90132fa43 --- /dev/null +++ b/checkpoint-200/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e229d60ec309072e519bf994d2598b033f557d6c3dcfdfbc6f6a9da8fefac82 +size 21687 diff --git a/checkpoint-200/rng_state_5.pth b/checkpoint-200/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..1544e33fa5503dc275de3677bc1b37821a48a6f6 --- /dev/null +++ b/checkpoint-200/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52648b0e03696bcd84ead3aad5a977af0f0ccc81eea894555b5affabb9e5a542 +size 21687 diff --git a/checkpoint-200/rng_state_6.pth b/checkpoint-200/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..af72562aebe501f967c30e2795108c0f5daea4bc --- /dev/null +++ b/checkpoint-200/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50a435a654f5aaea213bb4cafb22cb108c288369148ff58c005b32fef8b6c110 +size 21687 diff --git a/checkpoint-200/rng_state_7.pth b/checkpoint-200/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..acb7bb3d1faf671a327843e6772b4eb3a525eab4 --- /dev/null +++ b/checkpoint-200/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dab160dbcff51051e3a3bd97294a25c0a51a211996c74d8f398b5f4dbb00a6cd +size 21687 diff --git a/checkpoint-200/scheduler.pt b/checkpoint-200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..946f69e91ed1c5c9831c994d274775d814f7db04 --- /dev/null +++ b/checkpoint-200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e771c12d5841a508ccb39960c649085315e3de4b3cf131465739fa5742b41494 +size 627 diff --git a/checkpoint-200/trainer_state.json b/checkpoint-200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..cea5aff118ba0c9e308b1593628cd62e243b29af --- /dev/null +++ b/checkpoint-200/trainer_state.json @@ -0,0 +1,1251 @@ +{ + "best_metric": 1.477729082107544, + "best_model_checkpoint": "./pippa-sharegpt-13b-qlora/checkpoint-200", + "epoch": 2.1108179419525066, + "eval_steps": 50, + "global_step": 200, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 2e-05, + "loss": 1.9036, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 4e-05, + "loss": 1.9209, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 6e-05, + "loss": 1.9161, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 8e-05, + "loss": 1.8722, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 0.0001, + "loss": 1.866, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 0.00012, + "loss": 1.872, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 0.00014, + "loss": 1.9138, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 0.00016, + "loss": 1.8785, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 0.00018, + "loss": 1.8013, + "step": 9 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002, + "loss": 1.7859, + "step": 10 + }, + { + "epoch": 0.12, + "learning_rate": 0.00019999332998034514, + "loss": 1.7805, + "step": 11 + }, + { + "epoch": 0.13, + "learning_rate": 0.00019997332081116373, + "loss": 1.7716, + "step": 12 + }, + { + "epoch": 0.14, + "learning_rate": 0.00019993997516168689, + "loss": 1.7453, + "step": 13 + }, + { + "epoch": 0.15, + "learning_rate": 0.00019989329748023725, + "loss": 1.7306, + "step": 14 + }, + { + "epoch": 0.16, + "learning_rate": 0.00019983329399363598, + "loss": 1.7535, + "step": 15 + }, + { + "epoch": 0.17, + "learning_rate": 0.0001997599727063717, + "loss": 1.7811, + "step": 16 + }, + { + "epoch": 0.18, + "learning_rate": 0.000199673343399533, + "loss": 1.7455, + "step": 17 + }, + { + "epoch": 0.19, + "learning_rate": 0.00019957341762950344, + "loss": 1.748, + "step": 18 + }, + { + "epoch": 0.2, + "learning_rate": 0.0001994602087264201, + "loss": 1.723, + "step": 19 + }, + { + "epoch": 0.21, + "learning_rate": 0.00019933373179239502, + "loss": 1.7539, + "step": 20 + }, + { + "epoch": 0.22, + "learning_rate": 0.000199194003699501, + "loss": 1.7293, + "step": 21 + }, + { + "epoch": 0.23, + "learning_rate": 0.0001990410430875205, + "loss": 1.6857, + "step": 22 + }, + { + "epoch": 0.24, + "learning_rate": 0.0001988748703614594, + "loss": 1.7209, + "step": 23 + }, + { + "epoch": 0.25, + "learning_rate": 0.00019869550768882455, + "loss": 1.7127, + "step": 24 + }, + { + "epoch": 0.26, + "learning_rate": 0.0001985029789966671, + "loss": 1.7024, + "step": 25 + }, + { + "epoch": 0.27, + "learning_rate": 0.0001982973099683902, + "loss": 1.6927, + "step": 26 + }, + { + "epoch": 0.28, + "learning_rate": 0.00019807852804032305, + "loss": 1.6584, + "step": 27 + }, + { + "epoch": 0.3, + "learning_rate": 0.0001978466623980609, + "loss": 1.7149, + "step": 28 + }, + { + "epoch": 0.31, + "learning_rate": 0.00019760174397257156, + "loss": 1.6815, + "step": 29 + }, + { + "epoch": 0.32, + "learning_rate": 0.0001973438054360693, + "loss": 1.7067, + "step": 30 + }, + { + "epoch": 0.33, + "learning_rate": 0.00019707288119765623, + "loss": 1.6799, + "step": 31 + }, + { + "epoch": 0.34, + "learning_rate": 0.00019678900739873226, + "loss": 1.7206, + "step": 32 + }, + { + "epoch": 0.35, + "learning_rate": 0.0001964922219081738, + "loss": 1.6458, + "step": 33 + }, + { + "epoch": 0.36, + "learning_rate": 0.00019618256431728194, + "loss": 1.6649, + "step": 34 + }, + { + "epoch": 0.37, + "learning_rate": 0.00019586007593450097, + "loss": 1.6684, + "step": 35 + }, + { + "epoch": 0.38, + "learning_rate": 0.000195524799779908, + "loss": 1.7411, + "step": 36 + }, + { + "epoch": 0.39, + "learning_rate": 0.00019517678057947384, + "loss": 1.6678, + "step": 37 + }, + { + "epoch": 0.4, + "learning_rate": 0.0001948160647590966, + "loss": 1.6738, + "step": 38 + }, + { + "epoch": 0.41, + "learning_rate": 0.00019444270043840852, + "loss": 1.6804, + "step": 39 + }, + { + "epoch": 0.42, + "learning_rate": 0.00019405673742435678, + "loss": 1.6728, + "step": 40 + }, + { + "epoch": 0.43, + "learning_rate": 0.00019365822720455916, + "loss": 1.6941, + "step": 41 + }, + { + "epoch": 0.44, + "learning_rate": 0.00019324722294043558, + "loss": 1.6861, + "step": 42 + }, + { + "epoch": 0.45, + "learning_rate": 0.00019282377946011652, + "loss": 1.6751, + "step": 43 + }, + { + "epoch": 0.46, + "learning_rate": 0.0001923879532511287, + "loss": 1.7175, + "step": 44 + }, + { + "epoch": 0.47, + "learning_rate": 0.00019193980245285966, + "loss": 1.6454, + "step": 45 + }, + { + "epoch": 0.49, + "learning_rate": 0.0001914793868488021, + "loss": 1.6934, + "step": 46 + }, + { + "epoch": 0.5, + "learning_rate": 0.0001910067678585786, + "loss": 1.6534, + "step": 47 + }, + { + "epoch": 0.51, + "learning_rate": 0.00019052200852974819, + "loss": 1.6797, + "step": 48 + }, + { + "epoch": 0.52, + "learning_rate": 0.00019002517352939598, + "loss": 1.6566, + "step": 49 + }, + { + "epoch": 0.53, + "learning_rate": 0.00018951632913550626, + "loss": 1.657, + "step": 50 + }, + { + "epoch": 0.53, + "eval_loss": 1.653587818145752, + "eval_runtime": 200.2473, + "eval_samples_per_second": 6.672, + "eval_steps_per_second": 0.21, + "step": 50 + }, + { + "epoch": 0.54, + "learning_rate": 0.0001889955432281212, + "loss": 1.6736, + "step": 51 + }, + { + "epoch": 0.55, + "learning_rate": 0.00018846288528028555, + "loss": 1.6696, + "step": 52 + }, + { + "epoch": 0.56, + "learning_rate": 0.00018791842634877898, + "loss": 1.6429, + "step": 53 + }, + { + "epoch": 0.57, + "learning_rate": 0.00018736223906463696, + "loss": 1.608, + "step": 54 + }, + { + "epoch": 0.58, + "learning_rate": 0.00018679439762346185, + "loss": 1.6245, + "step": 55 + }, + { + "epoch": 0.59, + "learning_rate": 0.00018621497777552507, + "loss": 1.6199, + "step": 56 + }, + { + "epoch": 0.6, + "learning_rate": 0.00018562405681566216, + "loss": 1.7028, + "step": 57 + }, + { + "epoch": 0.61, + "learning_rate": 0.00018502171357296144, + "loss": 1.6372, + "step": 58 + }, + { + "epoch": 0.62, + "learning_rate": 0.00018440802840024822, + "loss": 1.6198, + "step": 59 + }, + { + "epoch": 0.63, + "learning_rate": 0.00018378308316336584, + "loss": 1.6061, + "step": 60 + }, + { + "epoch": 0.64, + "learning_rate": 0.00018314696123025454, + "loss": 1.6051, + "step": 61 + }, + { + "epoch": 0.65, + "learning_rate": 0.00018249974745983023, + "loss": 1.6257, + "step": 62 + }, + { + "epoch": 0.66, + "learning_rate": 0.00018184152819066435, + "loss": 1.6295, + "step": 63 + }, + { + "epoch": 0.68, + "learning_rate": 0.00018117239122946615, + "loss": 1.6382, + "step": 64 + }, + { + "epoch": 0.69, + "learning_rate": 0.0001804924258393692, + "loss": 1.6032, + "step": 65 + }, + { + "epoch": 0.7, + "learning_rate": 0.000179801722728024, + "loss": 1.652, + "step": 66 + }, + { + "epoch": 0.71, + "learning_rate": 0.00017910037403549693, + "loss": 1.6431, + "step": 67 + }, + { + "epoch": 0.72, + "learning_rate": 0.00017838847332197938, + "loss": 1.6482, + "step": 68 + }, + { + "epoch": 0.73, + "learning_rate": 0.00017766611555530636, + "loss": 1.6276, + "step": 69 + }, + { + "epoch": 0.74, + "learning_rate": 0.00017693339709828792, + "loss": 1.611, + "step": 70 + }, + { + "epoch": 0.75, + "learning_rate": 0.00017619041569585418, + "loss": 1.6309, + "step": 71 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001754372704620164, + "loss": 1.6316, + "step": 72 + }, + { + "epoch": 0.77, + "learning_rate": 0.00017467406186664474, + "loss": 1.5779, + "step": 73 + }, + { + "epoch": 0.78, + "learning_rate": 0.00017390089172206592, + "loss": 1.6015, + "step": 74 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001731178631694811, + "loss": 1.6133, + "step": 75 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017232508066520702, + "loss": 1.6048, + "step": 76 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017152264996674136, + "loss": 1.5792, + "step": 77 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017071067811865476, + "loss": 1.5756, + "step": 78 + }, + { + "epoch": 0.83, + "learning_rate": 0.00016988927343831095, + "loss": 1.5497, + "step": 79 + }, + { + "epoch": 0.84, + "learning_rate": 0.00016905854550141716, + "loss": 1.5604, + "step": 80 + }, + { + "epoch": 0.85, + "learning_rate": 0.00016821860512740671, + "loss": 1.5722, + "step": 81 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016736956436465573, + "loss": 1.5943, + "step": 82 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016651153647553567, + "loss": 1.538, + "step": 83 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016564463592130428, + "loss": 1.5875, + "step": 84 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001647689783468362, + "loss": 1.5835, + "step": 85 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016388468056519612, + "loss": 1.5724, + "step": 86 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016299186054205577, + "loss": 1.573, + "step": 87 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016209063737995715, + "loss": 1.543, + "step": 88 + }, + { + "epoch": 0.94, + "learning_rate": 0.00016118113130242432, + "loss": 1.5438, + "step": 89 + }, + { + "epoch": 0.95, + "learning_rate": 0.00016026346363792567, + "loss": 1.5711, + "step": 90 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015933775680368822, + "loss": 1.5542, + "step": 91 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015840413428936767, + "loss": 1.5209, + "step": 92 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001574627206405744, + "loss": 1.569, + "step": 93 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001565136414422592, + "loss": 1.5657, + "step": 94 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015555702330196023, + "loss": 1.5248, + "step": 95 + }, + { + "epoch": 1.01, + "learning_rate": 0.00015459299383291345, + "loss": 1.5482, + "step": 96 + }, + { + "epoch": 1.02, + "learning_rate": 0.000153621681637029, + "loss": 1.5186, + "step": 97 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001526432162877356, + "loss": 1.5363, + "step": 98 + }, + { + "epoch": 1.04, + "learning_rate": 0.00015165772831269547, + "loss": 1.5221, + "step": 99 + }, + { + "epoch": 1.06, + "learning_rate": 0.00015066534917639195, + "loss": 1.5071, + "step": 100 + }, + { + "epoch": 1.06, + "eval_loss": 1.541548728942871, + "eval_runtime": 200.1636, + "eval_samples_per_second": 6.675, + "eval_steps_per_second": 0.21, + "step": 100 + }, + { + "epoch": 1.07, + "learning_rate": 0.00014966621126259183, + "loss": 1.5179, + "step": 101 + }, + { + "epoch": 1.08, + "learning_rate": 0.00014866044785668563, + "loss": 1.4923, + "step": 102 + }, + { + "epoch": 1.09, + "learning_rate": 0.00014764819312790707, + "loss": 1.5733, + "step": 103 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001466295821114348, + "loss": 1.4586, + "step": 104 + }, + { + "epoch": 1.11, + "learning_rate": 0.00014560475069037894, + "loss": 1.5379, + "step": 105 + }, + { + "epoch": 1.12, + "learning_rate": 0.00014457383557765386, + "loss": 1.5059, + "step": 106 + }, + { + "epoch": 1.13, + "learning_rate": 0.00014353697429774084, + "loss": 1.512, + "step": 107 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001424943051683422, + "loss": 1.4908, + "step": 108 + }, + { + "epoch": 1.15, + "learning_rate": 0.0001414459672819297, + "loss": 1.5067, + "step": 109 + }, + { + "epoch": 1.16, + "learning_rate": 0.00014039210048718949, + "loss": 1.4908, + "step": 110 + }, + { + "epoch": 1.17, + "learning_rate": 0.00013933284537036625, + "loss": 1.5299, + "step": 111 + }, + { + "epoch": 1.18, + "learning_rate": 0.000138268343236509, + "loss": 1.4836, + "step": 112 + }, + { + "epoch": 1.19, + "learning_rate": 0.00013719873609062077, + "loss": 1.5285, + "step": 113 + }, + { + "epoch": 1.2, + "learning_rate": 0.00013612416661871533, + "loss": 1.5493, + "step": 114 + }, + { + "epoch": 1.21, + "learning_rate": 0.0001350447781687826, + "loss": 1.4915, + "step": 115 + }, + { + "epoch": 1.22, + "learning_rate": 0.00013396071473166613, + "loss": 1.5139, + "step": 116 + }, + { + "epoch": 1.23, + "learning_rate": 0.00013287212092185464, + "loss": 1.4589, + "step": 117 + }, + { + "epoch": 1.25, + "learning_rate": 0.00013177914195819016, + "loss": 1.4869, + "step": 118 + }, + { + "epoch": 1.26, + "learning_rate": 0.00013068192364449618, + "loss": 1.5081, + "step": 119 + }, + { + "epoch": 1.27, + "learning_rate": 0.00012958061235012706, + "loss": 1.5172, + "step": 120 + }, + { + "epoch": 1.28, + "learning_rate": 0.0001284753549904423, + "loss": 1.5195, + "step": 121 + }, + { + "epoch": 1.29, + "learning_rate": 0.0001273662990072083, + "loss": 1.4635, + "step": 122 + }, + { + "epoch": 1.3, + "learning_rate": 0.00012625359234892907, + "loss": 1.5036, + "step": 123 + }, + { + "epoch": 1.31, + "learning_rate": 0.0001251373834511103, + "loss": 1.5203, + "step": 124 + }, + { + "epoch": 1.32, + "learning_rate": 0.00012401782121645766, + "loss": 1.5047, + "step": 125 + }, + { + "epoch": 1.33, + "learning_rate": 0.0001228950549950134, + "loss": 1.5047, + "step": 126 + }, + { + "epoch": 1.34, + "learning_rate": 0.00012176923456423284, + "loss": 1.4715, + "step": 127 + }, + { + "epoch": 1.35, + "learning_rate": 0.00012064051010900397, + "loss": 1.4884, + "step": 128 + }, + { + "epoch": 1.36, + "learning_rate": 0.00011950903220161285, + "loss": 1.4947, + "step": 129 + }, + { + "epoch": 1.37, + "learning_rate": 0.00011837495178165706, + "loss": 1.4665, + "step": 130 + }, + { + "epoch": 1.38, + "learning_rate": 0.00011723842013591044, + "loss": 1.4888, + "step": 131 + }, + { + "epoch": 1.39, + "learning_rate": 0.00011609958887814129, + "loss": 1.52, + "step": 132 + }, + { + "epoch": 1.4, + "learning_rate": 0.00011495860992888712, + "loss": 1.4305, + "step": 133 + }, + { + "epoch": 1.41, + "learning_rate": 0.00011381563549518823, + "loss": 1.4865, + "step": 134 + }, + { + "epoch": 1.42, + "learning_rate": 0.00011267081805028339, + "loss": 1.5155, + "step": 135 + }, + { + "epoch": 1.44, + "learning_rate": 0.00011152431031326978, + "loss": 1.4952, + "step": 136 + }, + { + "epoch": 1.45, + "learning_rate": 0.00011037626522873019, + "loss": 1.4765, + "step": 137 + }, + { + "epoch": 1.46, + "learning_rate": 0.00010922683594633021, + "loss": 1.4729, + "step": 138 + }, + { + "epoch": 1.47, + "learning_rate": 0.00010807617580038796, + "loss": 1.481, + "step": 139 + }, + { + "epoch": 1.48, + "learning_rate": 0.00010692443828941918, + "loss": 1.4877, + "step": 140 + }, + { + "epoch": 1.49, + "learning_rate": 0.00010577177705566061, + "loss": 1.4884, + "step": 141 + }, + { + "epoch": 1.5, + "learning_rate": 0.00010461834586457398, + "loss": 1.4786, + "step": 142 + }, + { + "epoch": 1.51, + "learning_rate": 0.00010346429858433352, + "loss": 1.4585, + "step": 143 + }, + { + "epoch": 1.52, + "learning_rate": 0.00010230978916530012, + "loss": 1.4867, + "step": 144 + }, + { + "epoch": 1.53, + "learning_rate": 0.00010115497161948409, + "loss": 1.487, + "step": 145 + }, + { + "epoch": 1.54, + "learning_rate": 0.0001, + "loss": 1.4693, + "step": 146 + }, + { + "epoch": 1.55, + "learning_rate": 9.884502838051595e-05, + "loss": 1.4391, + "step": 147 + }, + { + "epoch": 1.56, + "learning_rate": 9.76902108346999e-05, + "loss": 1.493, + "step": 148 + }, + { + "epoch": 1.57, + "learning_rate": 9.653570141566653e-05, + "loss": 1.5152, + "step": 149 + }, + { + "epoch": 1.58, + "learning_rate": 9.538165413542607e-05, + "loss": 1.4766, + "step": 150 + }, + { + "epoch": 1.58, + "eval_loss": 1.4968100786209106, + "eval_runtime": 199.9343, + "eval_samples_per_second": 6.682, + "eval_steps_per_second": 0.21, + "step": 150 + }, + { + "epoch": 1.59, + "learning_rate": 9.422822294433939e-05, + "loss": 1.453, + "step": 151 + }, + { + "epoch": 1.6, + "learning_rate": 9.307556171058085e-05, + "loss": 1.5388, + "step": 152 + }, + { + "epoch": 1.61, + "learning_rate": 9.192382419961208e-05, + "loss": 1.5092, + "step": 153 + }, + { + "epoch": 1.63, + "learning_rate": 9.077316405366981e-05, + "loss": 1.4919, + "step": 154 + }, + { + "epoch": 1.64, + "learning_rate": 8.962373477126983e-05, + "loss": 1.4626, + "step": 155 + }, + { + "epoch": 1.65, + "learning_rate": 8.847568968673026e-05, + "loss": 1.4724, + "step": 156 + }, + { + "epoch": 1.66, + "learning_rate": 8.732918194971664e-05, + "loss": 1.4747, + "step": 157 + }, + { + "epoch": 1.67, + "learning_rate": 8.61843645048118e-05, + "loss": 1.4729, + "step": 158 + }, + { + "epoch": 1.68, + "learning_rate": 8.504139007111289e-05, + "loss": 1.4509, + "step": 159 + }, + { + "epoch": 1.69, + "learning_rate": 8.39004111218587e-05, + "loss": 1.457, + "step": 160 + }, + { + "epoch": 1.7, + "learning_rate": 8.27615798640896e-05, + "loss": 1.4104, + "step": 161 + }, + { + "epoch": 1.71, + "learning_rate": 8.162504821834295e-05, + "loss": 1.4846, + "step": 162 + }, + { + "epoch": 1.72, + "learning_rate": 8.049096779838719e-05, + "loss": 1.469, + "step": 163 + }, + { + "epoch": 1.73, + "learning_rate": 7.935948989099605e-05, + "loss": 1.501, + "step": 164 + }, + { + "epoch": 1.74, + "learning_rate": 7.823076543576717e-05, + "loss": 1.5113, + "step": 165 + }, + { + "epoch": 1.75, + "learning_rate": 7.710494500498662e-05, + "loss": 1.4365, + "step": 166 + }, + { + "epoch": 1.76, + "learning_rate": 7.598217878354237e-05, + "loss": 1.4719, + "step": 167 + }, + { + "epoch": 1.77, + "learning_rate": 7.486261654888973e-05, + "loss": 1.4694, + "step": 168 + }, + { + "epoch": 1.78, + "learning_rate": 7.374640765107095e-05, + "loss": 1.4342, + "step": 169 + }, + { + "epoch": 1.79, + "learning_rate": 7.263370099279172e-05, + "loss": 1.4961, + "step": 170 + }, + { + "epoch": 1.8, + "learning_rate": 7.152464500955769e-05, + "loss": 1.4992, + "step": 171 + }, + { + "epoch": 1.82, + "learning_rate": 7.041938764987297e-05, + "loss": 1.4588, + "step": 172 + }, + { + "epoch": 1.83, + "learning_rate": 6.931807635550383e-05, + "loss": 1.4351, + "step": 173 + }, + { + "epoch": 1.84, + "learning_rate": 6.822085804180984e-05, + "loss": 1.439, + "step": 174 + }, + { + "epoch": 1.85, + "learning_rate": 6.712787907814541e-05, + "loss": 1.5206, + "step": 175 + }, + { + "epoch": 1.86, + "learning_rate": 6.603928526833387e-05, + "loss": 1.4697, + "step": 176 + }, + { + "epoch": 1.87, + "learning_rate": 6.495522183121741e-05, + "loss": 1.509, + "step": 177 + }, + { + "epoch": 1.88, + "learning_rate": 6.387583338128471e-05, + "loss": 1.4888, + "step": 178 + }, + { + "epoch": 1.89, + "learning_rate": 6.280126390937925e-05, + "loss": 1.4322, + "step": 179 + }, + { + "epoch": 1.9, + "learning_rate": 6.173165676349103e-05, + "loss": 1.5042, + "step": 180 + }, + { + "epoch": 1.91, + "learning_rate": 6.066715462963376e-05, + "loss": 1.4575, + "step": 181 + }, + { + "epoch": 1.92, + "learning_rate": 5.960789951281052e-05, + "loss": 1.4351, + "step": 182 + }, + { + "epoch": 1.93, + "learning_rate": 5.855403271807033e-05, + "loss": 1.4421, + "step": 183 + }, + { + "epoch": 1.94, + "learning_rate": 5.750569483165784e-05, + "loss": 1.4899, + "step": 184 + }, + { + "epoch": 1.95, + "learning_rate": 5.646302570225919e-05, + "loss": 1.4524, + "step": 185 + }, + { + "epoch": 1.96, + "learning_rate": 5.542616442234618e-05, + "loss": 1.4322, + "step": 186 + }, + { + "epoch": 1.97, + "learning_rate": 5.43952493096211e-05, + "loss": 1.4829, + "step": 187 + }, + { + "epoch": 1.98, + "learning_rate": 5.337041788856518e-05, + "loss": 1.4034, + "step": 188 + }, + { + "epoch": 1.99, + "learning_rate": 5.235180687209296e-05, + "loss": 1.4996, + "step": 189 + }, + { + "epoch": 2.01, + "learning_rate": 5.1339552143314384e-05, + "loss": 1.4482, + "step": 190 + }, + { + "epoch": 2.02, + "learning_rate": 5.033378873740819e-05, + "loss": 1.3788, + "step": 191 + }, + { + "epoch": 2.03, + "learning_rate": 4.933465082360807e-05, + "loss": 1.4326, + "step": 192 + }, + { + "epoch": 2.04, + "learning_rate": 4.834227168730451e-05, + "loss": 1.4492, + "step": 193 + }, + { + "epoch": 2.05, + "learning_rate": 4.735678371226441e-05, + "loss": 1.4587, + "step": 194 + }, + { + "epoch": 2.06, + "learning_rate": 4.637831836297103e-05, + "loss": 1.4166, + "step": 195 + }, + { + "epoch": 2.07, + "learning_rate": 4.540700616708658e-05, + "loss": 1.4265, + "step": 196 + }, + { + "epoch": 2.08, + "learning_rate": 4.444297669803981e-05, + "loss": 1.4683, + "step": 197 + }, + { + "epoch": 2.09, + "learning_rate": 4.3486358557740814e-05, + "loss": 1.4413, + "step": 198 + }, + { + "epoch": 2.1, + "learning_rate": 4.253727935942563e-05, + "loss": 1.464, + "step": 199 + }, + { + "epoch": 2.11, + "learning_rate": 4.159586571063236e-05, + "loss": 1.4525, + "step": 200 + }, + { + "epoch": 2.11, + "eval_loss": 1.477729082107544, + "eval_runtime": 199.9883, + "eval_samples_per_second": 6.68, + "eval_steps_per_second": 0.21, + "step": 200 + } + ], + "logging_steps": 1, + "max_steps": 282, + "num_train_epochs": 3, + "save_steps": 50, + "total_flos": 2.070326304820953e+18, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-200/training_args.bin b/checkpoint-200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..41fe038f6bd77e41e63090845cfb2375d33e29ec --- /dev/null +++ b/checkpoint-200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8bdde5c34937c67d70adb2cea431eda05b8cf22eb5ebad425d871a634a8fc84 +size 4411 diff --git a/checkpoint-250/README.md b/checkpoint-250/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-250/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-250/adapter_config.json b/checkpoint-250/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-250/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-250/adapter_model.bin b/checkpoint-250/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..9d959aeaa72be12c730b665a117398f76bf4c7d6 --- /dev/null +++ b/checkpoint-250/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7738c4edc7b66e957a9a9e7474a3be85403c8e4f0c0a506d2806acd4fed40d5 +size 125374989 diff --git a/checkpoint-250/adapter_model/README.md b/checkpoint-250/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-250/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-250/adapter_model/adapter_config.json b/checkpoint-250/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-250/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-250/adapter_model/adapter_model.bin b/checkpoint-250/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..9d959aeaa72be12c730b665a117398f76bf4c7d6 --- /dev/null +++ b/checkpoint-250/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7738c4edc7b66e957a9a9e7474a3be85403c8e4f0c0a506d2806acd4fed40d5 +size 125374989 diff --git a/checkpoint-250/optimizer.pt b/checkpoint-250/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..ad67bfcf6b44655341c2d0e7c6363db82e79ff86 --- /dev/null +++ b/checkpoint-250/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fa83ff07b1b0765a322751ee6432ddb0dcb0b6a465a8c5b0ae99aeb4f82fef8 +size 250681597 diff --git a/checkpoint-250/rng_state_0.pth b/checkpoint-250/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..5f2eb580f1ecd8a54d3abc485e937d1982278ec0 --- /dev/null +++ b/checkpoint-250/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a141683fb3501481bcdee41542934d4abbe1448f4c01ad2406c1c6c78c18fd0a +size 21687 diff --git a/checkpoint-250/rng_state_1.pth b/checkpoint-250/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..383570cd8174b1153267e2263e52f55c35c59579 --- /dev/null +++ b/checkpoint-250/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0c8f3361fedc62dbd5925b84e8345ac628b36e5e58ba51f6823cae6018f53e6 +size 21687 diff --git a/checkpoint-250/rng_state_2.pth b/checkpoint-250/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..35b2eb6a22f14f2b0c78414579d289337b238e6c --- /dev/null +++ b/checkpoint-250/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7aad0d17dfd387e80953e99d57a0cf752cd95ad5532de5dd5a32bf039033deb +size 21687 diff --git a/checkpoint-250/rng_state_3.pth b/checkpoint-250/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..91b4cfc90d7cf7f3e63432aff333691d1e132641 --- /dev/null +++ b/checkpoint-250/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab2942eb4cc05e9c523e36f42ef10e7de139e257a6b3164bb820365413ce28c5 +size 21687 diff --git a/checkpoint-250/rng_state_4.pth b/checkpoint-250/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..9946aefcfcb685cb050516d8b0c1bb14b52e435b --- /dev/null +++ b/checkpoint-250/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cb1c4d46bcf0a98176f34b31b1e6416b3288d153f7432aad9a4b50193be2c11 +size 21687 diff --git a/checkpoint-250/rng_state_5.pth b/checkpoint-250/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..acf5ba4c1523376813261962d35a9b3e66011915 --- /dev/null +++ b/checkpoint-250/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01d1be61f011291a4cae8237ad40f9e29879d823cd47fa3c4b8c6f2e6d3c468e +size 21687 diff --git a/checkpoint-250/rng_state_6.pth b/checkpoint-250/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..23657f66d4212440f430cd6792c0f0686afbfb5b --- /dev/null +++ b/checkpoint-250/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb551ecad684889a3197f02db24045f003c4527c420a2763c2aca5ca1340e4f1 +size 21687 diff --git a/checkpoint-250/rng_state_7.pth b/checkpoint-250/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..ce8c7c139f66edc3c89d3f5b0bae36b78171f4ab --- /dev/null +++ b/checkpoint-250/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4c7f64985e921d3e6e56763bd872b6c1621f574909b8c74089ae27feac576b3 +size 21687 diff --git a/checkpoint-250/scheduler.pt b/checkpoint-250/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..7242d9c1adb5ee92efe045e5d8e40b78c7302514 --- /dev/null +++ b/checkpoint-250/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:727b5b8f7a3725edd38370e4f2b60c1788b09e64c341e52a850d058f72945282 +size 627 diff --git a/checkpoint-250/trainer_state.json b/checkpoint-250/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..aecdf9c6cbc2efe02a0dbda4232cc5eb1eedc3fa --- /dev/null +++ b/checkpoint-250/trainer_state.json @@ -0,0 +1,1559 @@ +{ + "best_metric": 1.4714999198913574, + "best_model_checkpoint": "./pippa-sharegpt-13b-qlora/checkpoint-250", + "epoch": 2.638522427440633, + "eval_steps": 50, + "global_step": 250, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 2e-05, + "loss": 1.9036, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 4e-05, + "loss": 1.9209, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 6e-05, + "loss": 1.9161, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 8e-05, + "loss": 1.8722, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 0.0001, + "loss": 1.866, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 0.00012, + "loss": 1.872, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 0.00014, + "loss": 1.9138, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 0.00016, + "loss": 1.8785, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 0.00018, + "loss": 1.8013, + "step": 9 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002, + "loss": 1.7859, + "step": 10 + }, + { + "epoch": 0.12, + "learning_rate": 0.00019999332998034514, + "loss": 1.7805, + "step": 11 + }, + { + "epoch": 0.13, + "learning_rate": 0.00019997332081116373, + "loss": 1.7716, + "step": 12 + }, + { + "epoch": 0.14, + "learning_rate": 0.00019993997516168689, + "loss": 1.7453, + "step": 13 + }, + { + "epoch": 0.15, + "learning_rate": 0.00019989329748023725, + "loss": 1.7306, + "step": 14 + }, + { + "epoch": 0.16, + "learning_rate": 0.00019983329399363598, + "loss": 1.7535, + "step": 15 + }, + { + "epoch": 0.17, + "learning_rate": 0.0001997599727063717, + "loss": 1.7811, + "step": 16 + }, + { + "epoch": 0.18, + "learning_rate": 0.000199673343399533, + "loss": 1.7455, + "step": 17 + }, + { + "epoch": 0.19, + "learning_rate": 0.00019957341762950344, + "loss": 1.748, + "step": 18 + }, + { + "epoch": 0.2, + "learning_rate": 0.0001994602087264201, + "loss": 1.723, + "step": 19 + }, + { + "epoch": 0.21, + "learning_rate": 0.00019933373179239502, + "loss": 1.7539, + "step": 20 + }, + { + "epoch": 0.22, + "learning_rate": 0.000199194003699501, + "loss": 1.7293, + "step": 21 + }, + { + "epoch": 0.23, + "learning_rate": 0.0001990410430875205, + "loss": 1.6857, + "step": 22 + }, + { + "epoch": 0.24, + "learning_rate": 0.0001988748703614594, + "loss": 1.7209, + "step": 23 + }, + { + "epoch": 0.25, + "learning_rate": 0.00019869550768882455, + "loss": 1.7127, + "step": 24 + }, + { + "epoch": 0.26, + "learning_rate": 0.0001985029789966671, + "loss": 1.7024, + "step": 25 + }, + { + "epoch": 0.27, + "learning_rate": 0.0001982973099683902, + "loss": 1.6927, + "step": 26 + }, + { + "epoch": 0.28, + "learning_rate": 0.00019807852804032305, + "loss": 1.6584, + "step": 27 + }, + { + "epoch": 0.3, + "learning_rate": 0.0001978466623980609, + "loss": 1.7149, + "step": 28 + }, + { + "epoch": 0.31, + "learning_rate": 0.00019760174397257156, + "loss": 1.6815, + "step": 29 + }, + { + "epoch": 0.32, + "learning_rate": 0.0001973438054360693, + "loss": 1.7067, + "step": 30 + }, + { + "epoch": 0.33, + "learning_rate": 0.00019707288119765623, + "loss": 1.6799, + "step": 31 + }, + { + "epoch": 0.34, + "learning_rate": 0.00019678900739873226, + "loss": 1.7206, + "step": 32 + }, + { + "epoch": 0.35, + "learning_rate": 0.0001964922219081738, + "loss": 1.6458, + "step": 33 + }, + { + "epoch": 0.36, + "learning_rate": 0.00019618256431728194, + "loss": 1.6649, + "step": 34 + }, + { + "epoch": 0.37, + "learning_rate": 0.00019586007593450097, + "loss": 1.6684, + "step": 35 + }, + { + "epoch": 0.38, + "learning_rate": 0.000195524799779908, + "loss": 1.7411, + "step": 36 + }, + { + "epoch": 0.39, + "learning_rate": 0.00019517678057947384, + "loss": 1.6678, + "step": 37 + }, + { + "epoch": 0.4, + "learning_rate": 0.0001948160647590966, + "loss": 1.6738, + "step": 38 + }, + { + "epoch": 0.41, + "learning_rate": 0.00019444270043840852, + "loss": 1.6804, + "step": 39 + }, + { + "epoch": 0.42, + "learning_rate": 0.00019405673742435678, + "loss": 1.6728, + "step": 40 + }, + { + "epoch": 0.43, + "learning_rate": 0.00019365822720455916, + "loss": 1.6941, + "step": 41 + }, + { + "epoch": 0.44, + "learning_rate": 0.00019324722294043558, + "loss": 1.6861, + "step": 42 + }, + { + "epoch": 0.45, + "learning_rate": 0.00019282377946011652, + "loss": 1.6751, + "step": 43 + }, + { + "epoch": 0.46, + "learning_rate": 0.0001923879532511287, + "loss": 1.7175, + "step": 44 + }, + { + "epoch": 0.47, + "learning_rate": 0.00019193980245285966, + "loss": 1.6454, + "step": 45 + }, + { + "epoch": 0.49, + "learning_rate": 0.0001914793868488021, + "loss": 1.6934, + "step": 46 + }, + { + "epoch": 0.5, + "learning_rate": 0.0001910067678585786, + "loss": 1.6534, + "step": 47 + }, + { + "epoch": 0.51, + "learning_rate": 0.00019052200852974819, + "loss": 1.6797, + "step": 48 + }, + { + "epoch": 0.52, + "learning_rate": 0.00019002517352939598, + "loss": 1.6566, + "step": 49 + }, + { + "epoch": 0.53, + "learning_rate": 0.00018951632913550626, + "loss": 1.657, + "step": 50 + }, + { + "epoch": 0.53, + "eval_loss": 1.653587818145752, + "eval_runtime": 200.2473, + "eval_samples_per_second": 6.672, + "eval_steps_per_second": 0.21, + "step": 50 + }, + { + "epoch": 0.54, + "learning_rate": 0.0001889955432281212, + "loss": 1.6736, + "step": 51 + }, + { + "epoch": 0.55, + "learning_rate": 0.00018846288528028555, + "loss": 1.6696, + "step": 52 + }, + { + "epoch": 0.56, + "learning_rate": 0.00018791842634877898, + "loss": 1.6429, + "step": 53 + }, + { + "epoch": 0.57, + "learning_rate": 0.00018736223906463696, + "loss": 1.608, + "step": 54 + }, + { + "epoch": 0.58, + "learning_rate": 0.00018679439762346185, + "loss": 1.6245, + "step": 55 + }, + { + "epoch": 0.59, + "learning_rate": 0.00018621497777552507, + "loss": 1.6199, + "step": 56 + }, + { + "epoch": 0.6, + "learning_rate": 0.00018562405681566216, + "loss": 1.7028, + "step": 57 + }, + { + "epoch": 0.61, + "learning_rate": 0.00018502171357296144, + "loss": 1.6372, + "step": 58 + }, + { + "epoch": 0.62, + "learning_rate": 0.00018440802840024822, + "loss": 1.6198, + "step": 59 + }, + { + "epoch": 0.63, + "learning_rate": 0.00018378308316336584, + "loss": 1.6061, + "step": 60 + }, + { + "epoch": 0.64, + "learning_rate": 0.00018314696123025454, + "loss": 1.6051, + "step": 61 + }, + { + "epoch": 0.65, + "learning_rate": 0.00018249974745983023, + "loss": 1.6257, + "step": 62 + }, + { + "epoch": 0.66, + "learning_rate": 0.00018184152819066435, + "loss": 1.6295, + "step": 63 + }, + { + "epoch": 0.68, + "learning_rate": 0.00018117239122946615, + "loss": 1.6382, + "step": 64 + }, + { + "epoch": 0.69, + "learning_rate": 0.0001804924258393692, + "loss": 1.6032, + "step": 65 + }, + { + "epoch": 0.7, + "learning_rate": 0.000179801722728024, + "loss": 1.652, + "step": 66 + }, + { + "epoch": 0.71, + "learning_rate": 0.00017910037403549693, + "loss": 1.6431, + "step": 67 + }, + { + "epoch": 0.72, + "learning_rate": 0.00017838847332197938, + "loss": 1.6482, + "step": 68 + }, + { + "epoch": 0.73, + "learning_rate": 0.00017766611555530636, + "loss": 1.6276, + "step": 69 + }, + { + "epoch": 0.74, + "learning_rate": 0.00017693339709828792, + "loss": 1.611, + "step": 70 + }, + { + "epoch": 0.75, + "learning_rate": 0.00017619041569585418, + "loss": 1.6309, + "step": 71 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001754372704620164, + "loss": 1.6316, + "step": 72 + }, + { + "epoch": 0.77, + "learning_rate": 0.00017467406186664474, + "loss": 1.5779, + "step": 73 + }, + { + "epoch": 0.78, + "learning_rate": 0.00017390089172206592, + "loss": 1.6015, + "step": 74 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001731178631694811, + "loss": 1.6133, + "step": 75 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017232508066520702, + "loss": 1.6048, + "step": 76 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017152264996674136, + "loss": 1.5792, + "step": 77 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017071067811865476, + "loss": 1.5756, + "step": 78 + }, + { + "epoch": 0.83, + "learning_rate": 0.00016988927343831095, + "loss": 1.5497, + "step": 79 + }, + { + "epoch": 0.84, + "learning_rate": 0.00016905854550141716, + "loss": 1.5604, + "step": 80 + }, + { + "epoch": 0.85, + "learning_rate": 0.00016821860512740671, + "loss": 1.5722, + "step": 81 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016736956436465573, + "loss": 1.5943, + "step": 82 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016651153647553567, + "loss": 1.538, + "step": 83 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016564463592130428, + "loss": 1.5875, + "step": 84 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001647689783468362, + "loss": 1.5835, + "step": 85 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016388468056519612, + "loss": 1.5724, + "step": 86 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016299186054205577, + "loss": 1.573, + "step": 87 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016209063737995715, + "loss": 1.543, + "step": 88 + }, + { + "epoch": 0.94, + "learning_rate": 0.00016118113130242432, + "loss": 1.5438, + "step": 89 + }, + { + "epoch": 0.95, + "learning_rate": 0.00016026346363792567, + "loss": 1.5711, + "step": 90 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015933775680368822, + "loss": 1.5542, + "step": 91 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015840413428936767, + "loss": 1.5209, + "step": 92 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001574627206405744, + "loss": 1.569, + "step": 93 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001565136414422592, + "loss": 1.5657, + "step": 94 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015555702330196023, + "loss": 1.5248, + "step": 95 + }, + { + "epoch": 1.01, + "learning_rate": 0.00015459299383291345, + "loss": 1.5482, + "step": 96 + }, + { + "epoch": 1.02, + "learning_rate": 0.000153621681637029, + "loss": 1.5186, + "step": 97 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001526432162877356, + "loss": 1.5363, + "step": 98 + }, + { + "epoch": 1.04, + "learning_rate": 0.00015165772831269547, + "loss": 1.5221, + "step": 99 + }, + { + "epoch": 1.06, + "learning_rate": 0.00015066534917639195, + "loss": 1.5071, + "step": 100 + }, + { + "epoch": 1.06, + "eval_loss": 1.541548728942871, + "eval_runtime": 200.1636, + "eval_samples_per_second": 6.675, + "eval_steps_per_second": 0.21, + "step": 100 + }, + { + "epoch": 1.07, + "learning_rate": 0.00014966621126259183, + "loss": 1.5179, + "step": 101 + }, + { + "epoch": 1.08, + "learning_rate": 0.00014866044785668563, + "loss": 1.4923, + "step": 102 + }, + { + "epoch": 1.09, + "learning_rate": 0.00014764819312790707, + "loss": 1.5733, + "step": 103 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001466295821114348, + "loss": 1.4586, + "step": 104 + }, + { + "epoch": 1.11, + "learning_rate": 0.00014560475069037894, + "loss": 1.5379, + "step": 105 + }, + { + "epoch": 1.12, + "learning_rate": 0.00014457383557765386, + "loss": 1.5059, + "step": 106 + }, + { + "epoch": 1.13, + "learning_rate": 0.00014353697429774084, + "loss": 1.512, + "step": 107 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001424943051683422, + "loss": 1.4908, + "step": 108 + }, + { + "epoch": 1.15, + "learning_rate": 0.0001414459672819297, + "loss": 1.5067, + "step": 109 + }, + { + "epoch": 1.16, + "learning_rate": 0.00014039210048718949, + "loss": 1.4908, + "step": 110 + }, + { + "epoch": 1.17, + "learning_rate": 0.00013933284537036625, + "loss": 1.5299, + "step": 111 + }, + { + "epoch": 1.18, + "learning_rate": 0.000138268343236509, + "loss": 1.4836, + "step": 112 + }, + { + "epoch": 1.19, + "learning_rate": 0.00013719873609062077, + "loss": 1.5285, + "step": 113 + }, + { + "epoch": 1.2, + "learning_rate": 0.00013612416661871533, + "loss": 1.5493, + "step": 114 + }, + { + "epoch": 1.21, + "learning_rate": 0.0001350447781687826, + "loss": 1.4915, + "step": 115 + }, + { + "epoch": 1.22, + "learning_rate": 0.00013396071473166613, + "loss": 1.5139, + "step": 116 + }, + { + "epoch": 1.23, + "learning_rate": 0.00013287212092185464, + "loss": 1.4589, + "step": 117 + }, + { + "epoch": 1.25, + "learning_rate": 0.00013177914195819016, + "loss": 1.4869, + "step": 118 + }, + { + "epoch": 1.26, + "learning_rate": 0.00013068192364449618, + "loss": 1.5081, + "step": 119 + }, + { + "epoch": 1.27, + "learning_rate": 0.00012958061235012706, + "loss": 1.5172, + "step": 120 + }, + { + "epoch": 1.28, + "learning_rate": 0.0001284753549904423, + "loss": 1.5195, + "step": 121 + }, + { + "epoch": 1.29, + "learning_rate": 0.0001273662990072083, + "loss": 1.4635, + "step": 122 + }, + { + "epoch": 1.3, + "learning_rate": 0.00012625359234892907, + "loss": 1.5036, + "step": 123 + }, + { + "epoch": 1.31, + "learning_rate": 0.0001251373834511103, + "loss": 1.5203, + "step": 124 + }, + { + "epoch": 1.32, + "learning_rate": 0.00012401782121645766, + "loss": 1.5047, + "step": 125 + }, + { + "epoch": 1.33, + "learning_rate": 0.0001228950549950134, + "loss": 1.5047, + "step": 126 + }, + { + "epoch": 1.34, + "learning_rate": 0.00012176923456423284, + "loss": 1.4715, + "step": 127 + }, + { + "epoch": 1.35, + "learning_rate": 0.00012064051010900397, + "loss": 1.4884, + "step": 128 + }, + { + "epoch": 1.36, + "learning_rate": 0.00011950903220161285, + "loss": 1.4947, + "step": 129 + }, + { + "epoch": 1.37, + "learning_rate": 0.00011837495178165706, + "loss": 1.4665, + "step": 130 + }, + { + "epoch": 1.38, + "learning_rate": 0.00011723842013591044, + "loss": 1.4888, + "step": 131 + }, + { + "epoch": 1.39, + "learning_rate": 0.00011609958887814129, + "loss": 1.52, + "step": 132 + }, + { + "epoch": 1.4, + "learning_rate": 0.00011495860992888712, + "loss": 1.4305, + "step": 133 + }, + { + "epoch": 1.41, + "learning_rate": 0.00011381563549518823, + "loss": 1.4865, + "step": 134 + }, + { + "epoch": 1.42, + "learning_rate": 0.00011267081805028339, + "loss": 1.5155, + "step": 135 + }, + { + "epoch": 1.44, + "learning_rate": 0.00011152431031326978, + "loss": 1.4952, + "step": 136 + }, + { + "epoch": 1.45, + "learning_rate": 0.00011037626522873019, + "loss": 1.4765, + "step": 137 + }, + { + "epoch": 1.46, + "learning_rate": 0.00010922683594633021, + "loss": 1.4729, + "step": 138 + }, + { + "epoch": 1.47, + "learning_rate": 0.00010807617580038796, + "loss": 1.481, + "step": 139 + }, + { + "epoch": 1.48, + "learning_rate": 0.00010692443828941918, + "loss": 1.4877, + "step": 140 + }, + { + "epoch": 1.49, + "learning_rate": 0.00010577177705566061, + "loss": 1.4884, + "step": 141 + }, + { + "epoch": 1.5, + "learning_rate": 0.00010461834586457398, + "loss": 1.4786, + "step": 142 + }, + { + "epoch": 1.51, + "learning_rate": 0.00010346429858433352, + "loss": 1.4585, + "step": 143 + }, + { + "epoch": 1.52, + "learning_rate": 0.00010230978916530012, + "loss": 1.4867, + "step": 144 + }, + { + "epoch": 1.53, + "learning_rate": 0.00010115497161948409, + "loss": 1.487, + "step": 145 + }, + { + "epoch": 1.54, + "learning_rate": 0.0001, + "loss": 1.4693, + "step": 146 + }, + { + "epoch": 1.55, + "learning_rate": 9.884502838051595e-05, + "loss": 1.4391, + "step": 147 + }, + { + "epoch": 1.56, + "learning_rate": 9.76902108346999e-05, + "loss": 1.493, + "step": 148 + }, + { + "epoch": 1.57, + "learning_rate": 9.653570141566653e-05, + "loss": 1.5152, + "step": 149 + }, + { + "epoch": 1.58, + "learning_rate": 9.538165413542607e-05, + "loss": 1.4766, + "step": 150 + }, + { + "epoch": 1.58, + "eval_loss": 1.4968100786209106, + "eval_runtime": 199.9343, + "eval_samples_per_second": 6.682, + "eval_steps_per_second": 0.21, + "step": 150 + }, + { + "epoch": 1.59, + "learning_rate": 9.422822294433939e-05, + "loss": 1.453, + "step": 151 + }, + { + "epoch": 1.6, + "learning_rate": 9.307556171058085e-05, + "loss": 1.5388, + "step": 152 + }, + { + "epoch": 1.61, + "learning_rate": 9.192382419961208e-05, + "loss": 1.5092, + "step": 153 + }, + { + "epoch": 1.63, + "learning_rate": 9.077316405366981e-05, + "loss": 1.4919, + "step": 154 + }, + { + "epoch": 1.64, + "learning_rate": 8.962373477126983e-05, + "loss": 1.4626, + "step": 155 + }, + { + "epoch": 1.65, + "learning_rate": 8.847568968673026e-05, + "loss": 1.4724, + "step": 156 + }, + { + "epoch": 1.66, + "learning_rate": 8.732918194971664e-05, + "loss": 1.4747, + "step": 157 + }, + { + "epoch": 1.67, + "learning_rate": 8.61843645048118e-05, + "loss": 1.4729, + "step": 158 + }, + { + "epoch": 1.68, + "learning_rate": 8.504139007111289e-05, + "loss": 1.4509, + "step": 159 + }, + { + "epoch": 1.69, + "learning_rate": 8.39004111218587e-05, + "loss": 1.457, + "step": 160 + }, + { + "epoch": 1.7, + "learning_rate": 8.27615798640896e-05, + "loss": 1.4104, + "step": 161 + }, + { + "epoch": 1.71, + "learning_rate": 8.162504821834295e-05, + "loss": 1.4846, + "step": 162 + }, + { + "epoch": 1.72, + "learning_rate": 8.049096779838719e-05, + "loss": 1.469, + "step": 163 + }, + { + "epoch": 1.73, + "learning_rate": 7.935948989099605e-05, + "loss": 1.501, + "step": 164 + }, + { + "epoch": 1.74, + "learning_rate": 7.823076543576717e-05, + "loss": 1.5113, + "step": 165 + }, + { + "epoch": 1.75, + "learning_rate": 7.710494500498662e-05, + "loss": 1.4365, + "step": 166 + }, + { + "epoch": 1.76, + "learning_rate": 7.598217878354237e-05, + "loss": 1.4719, + "step": 167 + }, + { + "epoch": 1.77, + "learning_rate": 7.486261654888973e-05, + "loss": 1.4694, + "step": 168 + }, + { + "epoch": 1.78, + "learning_rate": 7.374640765107095e-05, + "loss": 1.4342, + "step": 169 + }, + { + "epoch": 1.79, + "learning_rate": 7.263370099279172e-05, + "loss": 1.4961, + "step": 170 + }, + { + "epoch": 1.8, + "learning_rate": 7.152464500955769e-05, + "loss": 1.4992, + "step": 171 + }, + { + "epoch": 1.82, + "learning_rate": 7.041938764987297e-05, + "loss": 1.4588, + "step": 172 + }, + { + "epoch": 1.83, + "learning_rate": 6.931807635550383e-05, + "loss": 1.4351, + "step": 173 + }, + { + "epoch": 1.84, + "learning_rate": 6.822085804180984e-05, + "loss": 1.439, + "step": 174 + }, + { + "epoch": 1.85, + "learning_rate": 6.712787907814541e-05, + "loss": 1.5206, + "step": 175 + }, + { + "epoch": 1.86, + "learning_rate": 6.603928526833387e-05, + "loss": 1.4697, + "step": 176 + }, + { + "epoch": 1.87, + "learning_rate": 6.495522183121741e-05, + "loss": 1.509, + "step": 177 + }, + { + "epoch": 1.88, + "learning_rate": 6.387583338128471e-05, + "loss": 1.4888, + "step": 178 + }, + { + "epoch": 1.89, + "learning_rate": 6.280126390937925e-05, + "loss": 1.4322, + "step": 179 + }, + { + "epoch": 1.9, + "learning_rate": 6.173165676349103e-05, + "loss": 1.5042, + "step": 180 + }, + { + "epoch": 1.91, + "learning_rate": 6.066715462963376e-05, + "loss": 1.4575, + "step": 181 + }, + { + "epoch": 1.92, + "learning_rate": 5.960789951281052e-05, + "loss": 1.4351, + "step": 182 + }, + { + "epoch": 1.93, + "learning_rate": 5.855403271807033e-05, + "loss": 1.4421, + "step": 183 + }, + { + "epoch": 1.94, + "learning_rate": 5.750569483165784e-05, + "loss": 1.4899, + "step": 184 + }, + { + "epoch": 1.95, + "learning_rate": 5.646302570225919e-05, + "loss": 1.4524, + "step": 185 + }, + { + "epoch": 1.96, + "learning_rate": 5.542616442234618e-05, + "loss": 1.4322, + "step": 186 + }, + { + "epoch": 1.97, + "learning_rate": 5.43952493096211e-05, + "loss": 1.4829, + "step": 187 + }, + { + "epoch": 1.98, + "learning_rate": 5.337041788856518e-05, + "loss": 1.4034, + "step": 188 + }, + { + "epoch": 1.99, + "learning_rate": 5.235180687209296e-05, + "loss": 1.4996, + "step": 189 + }, + { + "epoch": 2.01, + "learning_rate": 5.1339552143314384e-05, + "loss": 1.4482, + "step": 190 + }, + { + "epoch": 2.02, + "learning_rate": 5.033378873740819e-05, + "loss": 1.3788, + "step": 191 + }, + { + "epoch": 2.03, + "learning_rate": 4.933465082360807e-05, + "loss": 1.4326, + "step": 192 + }, + { + "epoch": 2.04, + "learning_rate": 4.834227168730451e-05, + "loss": 1.4492, + "step": 193 + }, + { + "epoch": 2.05, + "learning_rate": 4.735678371226441e-05, + "loss": 1.4587, + "step": 194 + }, + { + "epoch": 2.06, + "learning_rate": 4.637831836297103e-05, + "loss": 1.4166, + "step": 195 + }, + { + "epoch": 2.07, + "learning_rate": 4.540700616708658e-05, + "loss": 1.4265, + "step": 196 + }, + { + "epoch": 2.08, + "learning_rate": 4.444297669803981e-05, + "loss": 1.4683, + "step": 197 + }, + { + "epoch": 2.09, + "learning_rate": 4.3486358557740814e-05, + "loss": 1.4413, + "step": 198 + }, + { + "epoch": 2.1, + "learning_rate": 4.253727935942563e-05, + "loss": 1.464, + "step": 199 + }, + { + "epoch": 2.11, + "learning_rate": 4.159586571063236e-05, + "loss": 1.4525, + "step": 200 + }, + { + "epoch": 2.11, + "eval_loss": 1.477729082107544, + "eval_runtime": 199.9883, + "eval_samples_per_second": 6.68, + "eval_steps_per_second": 0.21, + "step": 200 + }, + { + "epoch": 2.12, + "learning_rate": 4.0662243196311815e-05, + "loss": 1.4009, + "step": 201 + }, + { + "epoch": 2.13, + "learning_rate": 3.973653636207437e-05, + "loss": 1.3728, + "step": 202 + }, + { + "epoch": 2.14, + "learning_rate": 3.881886869757565e-05, + "loss": 1.4354, + "step": 203 + }, + { + "epoch": 2.15, + "learning_rate": 3.7909362620042865e-05, + "loss": 1.4239, + "step": 204 + }, + { + "epoch": 2.16, + "learning_rate": 3.7008139457944245e-05, + "loss": 1.4413, + "step": 205 + }, + { + "epoch": 2.17, + "learning_rate": 3.6115319434803894e-05, + "loss": 1.3973, + "step": 206 + }, + { + "epoch": 2.18, + "learning_rate": 3.523102165316381e-05, + "loss": 1.4753, + "step": 207 + }, + { + "epoch": 2.2, + "learning_rate": 3.435536407869575e-05, + "loss": 1.4484, + "step": 208 + }, + { + "epoch": 2.21, + "learning_rate": 3.348846352446435e-05, + "loss": 1.3755, + "step": 209 + }, + { + "epoch": 2.22, + "learning_rate": 3.263043563534428e-05, + "loss": 1.3889, + "step": 210 + }, + { + "epoch": 2.23, + "learning_rate": 3.1781394872593295e-05, + "loss": 1.4591, + "step": 211 + }, + { + "epoch": 2.24, + "learning_rate": 3.094145449858285e-05, + "loss": 1.4091, + "step": 212 + }, + { + "epoch": 2.25, + "learning_rate": 3.0110726561689063e-05, + "loss": 1.4553, + "step": 213 + }, + { + "epoch": 2.26, + "learning_rate": 2.9289321881345254e-05, + "loss": 1.4053, + "step": 214 + }, + { + "epoch": 2.27, + "learning_rate": 2.8477350033258677e-05, + "loss": 1.507, + "step": 215 + }, + { + "epoch": 2.28, + "learning_rate": 2.7674919334793035e-05, + "loss": 1.4051, + "step": 216 + }, + { + "epoch": 2.29, + "learning_rate": 2.688213683051892e-05, + "loss": 1.4654, + "step": 217 + }, + { + "epoch": 2.3, + "learning_rate": 2.6099108277934103e-05, + "loss": 1.4758, + "step": 218 + }, + { + "epoch": 2.31, + "learning_rate": 2.5325938133355242e-05, + "loss": 1.4583, + "step": 219 + }, + { + "epoch": 2.32, + "learning_rate": 2.456272953798361e-05, + "loss": 1.446, + "step": 220 + }, + { + "epoch": 2.33, + "learning_rate": 2.3809584304145827e-05, + "loss": 1.4289, + "step": 221 + }, + { + "epoch": 2.34, + "learning_rate": 2.3066602901712108e-05, + "loss": 1.4231, + "step": 222 + }, + { + "epoch": 2.35, + "learning_rate": 2.2333884444693652e-05, + "loss": 1.4555, + "step": 223 + }, + { + "epoch": 2.36, + "learning_rate": 2.1611526678020654e-05, + "loss": 1.3931, + "step": 224 + }, + { + "epoch": 2.37, + "learning_rate": 2.0899625964503113e-05, + "loss": 1.4323, + "step": 225 + }, + { + "epoch": 2.39, + "learning_rate": 2.0198277271976052e-05, + "loss": 1.4633, + "step": 226 + }, + { + "epoch": 2.4, + "learning_rate": 1.950757416063077e-05, + "loss": 1.4225, + "step": 227 + }, + { + "epoch": 2.41, + "learning_rate": 1.882760877053388e-05, + "loss": 1.4635, + "step": 228 + }, + { + "epoch": 2.42, + "learning_rate": 1.8158471809335654e-05, + "loss": 1.4587, + "step": 229 + }, + { + "epoch": 2.43, + "learning_rate": 1.750025254016978e-05, + "loss": 1.4445, + "step": 230 + }, + { + "epoch": 2.44, + "learning_rate": 1.6853038769745467e-05, + "loss": 1.4704, + "step": 231 + }, + { + "epoch": 2.45, + "learning_rate": 1.621691683663418e-05, + "loss": 1.4673, + "step": 232 + }, + { + "epoch": 2.46, + "learning_rate": 1.55919715997518e-05, + "loss": 1.4398, + "step": 233 + }, + { + "epoch": 2.47, + "learning_rate": 1.4978286427038601e-05, + "loss": 1.4683, + "step": 234 + }, + { + "epoch": 2.48, + "learning_rate": 1.437594318433787e-05, + "loss": 1.386, + "step": 235 + }, + { + "epoch": 2.49, + "learning_rate": 1.3785022224474942e-05, + "loss": 1.4151, + "step": 236 + }, + { + "epoch": 2.5, + "learning_rate": 1.3205602376538163e-05, + "loss": 1.3905, + "step": 237 + }, + { + "epoch": 2.51, + "learning_rate": 1.2637760935363053e-05, + "loss": 1.4065, + "step": 238 + }, + { + "epoch": 2.52, + "learning_rate": 1.2081573651221035e-05, + "loss": 1.4331, + "step": 239 + }, + { + "epoch": 2.53, + "learning_rate": 1.1537114719714481e-05, + "loss": 1.4142, + "step": 240 + }, + { + "epoch": 2.54, + "learning_rate": 1.1004456771878835e-05, + "loss": 1.4813, + "step": 241 + }, + { + "epoch": 2.55, + "learning_rate": 1.0483670864493778e-05, + "loss": 1.4338, + "step": 242 + }, + { + "epoch": 2.56, + "learning_rate": 9.974826470604049e-06, + "loss": 1.4338, + "step": 243 + }, + { + "epoch": 2.58, + "learning_rate": 9.477991470251791e-06, + "loss": 1.4079, + "step": 244 + }, + { + "epoch": 2.59, + "learning_rate": 8.993232141421415e-06, + "loss": 1.3986, + "step": 245 + }, + { + "epoch": 2.6, + "learning_rate": 8.520613151197898e-06, + "loss": 1.4242, + "step": 246 + }, + { + "epoch": 2.61, + "learning_rate": 8.060197547140347e-06, + "loss": 1.4173, + "step": 247 + }, + { + "epoch": 2.62, + "learning_rate": 7.612046748871327e-06, + "loss": 1.456, + "step": 248 + }, + { + "epoch": 2.63, + "learning_rate": 7.1762205398834936e-06, + "loss": 1.4633, + "step": 249 + }, + { + "epoch": 2.64, + "learning_rate": 6.75277705956443e-06, + "loss": 1.4515, + "step": 250 + }, + { + "epoch": 2.64, + "eval_loss": 1.4714999198913574, + "eval_runtime": 200.1014, + "eval_samples_per_second": 6.677, + "eval_steps_per_second": 0.21, + "step": 250 + } + ], + "logging_steps": 1, + "max_steps": 282, + "num_train_epochs": 3, + "save_steps": 50, + "total_flos": 2.581735490512224e+18, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-250/training_args.bin b/checkpoint-250/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..41fe038f6bd77e41e63090845cfb2375d33e29ec --- /dev/null +++ b/checkpoint-250/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8bdde5c34937c67d70adb2cea431eda05b8cf22eb5ebad425d871a634a8fc84 +size 4411 diff --git a/checkpoint-50/README.md b/checkpoint-50/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-50/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-50/adapter_config.json b/checkpoint-50/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-50/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-50/adapter_model.bin b/checkpoint-50/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..d87f018c55b670d63980603751eadcc8ef3dbb8f --- /dev/null +++ b/checkpoint-50/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e57d4ffe33339139720eaa4613473810fea85e3e4d8653688a8dfa5bc8f7b02d +size 125374989 diff --git a/checkpoint-50/adapter_model/README.md b/checkpoint-50/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/checkpoint-50/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/checkpoint-50/adapter_model/adapter_config.json b/checkpoint-50/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3aec33e63c62e54f17eaebabde8f14c1805c7a13 --- /dev/null +++ b/checkpoint-50/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.05, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "k_proj", + "up_proj", + "o_proj", + "gate_proj", + "q_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/checkpoint-50/adapter_model/adapter_model.bin b/checkpoint-50/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..d87f018c55b670d63980603751eadcc8ef3dbb8f --- /dev/null +++ b/checkpoint-50/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e57d4ffe33339139720eaa4613473810fea85e3e4d8653688a8dfa5bc8f7b02d +size 125374989 diff --git a/checkpoint-50/optimizer.pt b/checkpoint-50/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..e8726c561ed3c93d1e5a91e535df4829cbd3267b --- /dev/null +++ b/checkpoint-50/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3769f3bc8b724fa5b2f46a6cac3d5cd53afaee13cbe7aac9e232201ef772966e +size 250681597 diff --git a/checkpoint-50/rng_state_0.pth b/checkpoint-50/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..607173aadfda9cf7c074c62788d35f61394a9935 --- /dev/null +++ b/checkpoint-50/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5603696c9f025e85d331d23aef42a578c026c834e7a5902b2fcee441625cc412 +size 21687 diff --git a/checkpoint-50/rng_state_1.pth b/checkpoint-50/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..e3c37c69df2018ca014f0f4fbb984262e4bbd42d --- /dev/null +++ b/checkpoint-50/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f141cab09b94890fd22d75d6af0c801d0582e3f7ef409cb7c576f08c96a7f21 +size 21687 diff --git a/checkpoint-50/rng_state_2.pth b/checkpoint-50/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..7dadcc7f2b6a1609b173d88e0f31034a51751f31 --- /dev/null +++ b/checkpoint-50/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a1ab6b4935ee34ae3ee9169e70a1d00b4e00c1ff1ad0c89b54e3aafb63302fb +size 21687 diff --git a/checkpoint-50/rng_state_3.pth b/checkpoint-50/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..e8ffcf924513c1bd426d17ca915b3c3eb1f220e6 --- /dev/null +++ b/checkpoint-50/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70493e263fa181ae47ea40f0e7e88444958444e76c2e92929afaccdb001c82b7 +size 21687 diff --git a/checkpoint-50/rng_state_4.pth b/checkpoint-50/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..1513782027fb462af32266976f9c4985d52b770b --- /dev/null +++ b/checkpoint-50/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e32be8436ba8e27fee6b4a7e7812cb6cc36771f3064a998547309975486537d +size 21687 diff --git a/checkpoint-50/rng_state_5.pth b/checkpoint-50/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..56de8b773e8ce91a2d27f29863c0ce9d8062bcfa --- /dev/null +++ b/checkpoint-50/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07d846c25348af6c0a87df59fa8b1972f8fa3995b3f9d2691eb4366a0e24f15a +size 21687 diff --git a/checkpoint-50/rng_state_6.pth b/checkpoint-50/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..16ce492cda1f022338f77d6145df7d49cb32ef2f --- /dev/null +++ b/checkpoint-50/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db2f85f6b573d33c7473ef85706c175693a0ddd7013edf4635e888b2697a25e6 +size 21687 diff --git a/checkpoint-50/rng_state_7.pth b/checkpoint-50/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..3e91b904262c138c7a2cc5a00641b2a34aaa13f1 --- /dev/null +++ b/checkpoint-50/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:556f381c274591fb8038bd4a14a9c452e25347f8f27b928069893af8a1bb56aa +size 21687 diff --git a/checkpoint-50/scheduler.pt b/checkpoint-50/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..9b0651f3903fcde8e6ec7aed2f2e5310018accc1 --- /dev/null +++ b/checkpoint-50/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52a1998b3ce3c33a63b2e0c342014b9ddbfc510ba4cbf41a6fe6c92dccadc89b +size 627 diff --git a/checkpoint-50/trainer_state.json b/checkpoint-50/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8344dce73db8472ef829aafe231fc9780a59d9bd --- /dev/null +++ b/checkpoint-50/trainer_state.json @@ -0,0 +1,327 @@ +{ + "best_metric": 1.653587818145752, + "best_model_checkpoint": "./pippa-sharegpt-13b-qlora/checkpoint-50", + "epoch": 0.5277044854881267, + "eval_steps": 50, + "global_step": 50, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 2e-05, + "loss": 1.9036, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 4e-05, + "loss": 1.9209, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 6e-05, + "loss": 1.9161, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 8e-05, + "loss": 1.8722, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 0.0001, + "loss": 1.866, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 0.00012, + "loss": 1.872, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 0.00014, + "loss": 1.9138, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 0.00016, + "loss": 1.8785, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 0.00018, + "loss": 1.8013, + "step": 9 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002, + "loss": 1.7859, + "step": 10 + }, + { + "epoch": 0.12, + "learning_rate": 0.00019999332998034514, + "loss": 1.7805, + "step": 11 + }, + { + "epoch": 0.13, + "learning_rate": 0.00019997332081116373, + "loss": 1.7716, + "step": 12 + }, + { + "epoch": 0.14, + "learning_rate": 0.00019993997516168689, + "loss": 1.7453, + "step": 13 + }, + { + "epoch": 0.15, + "learning_rate": 0.00019989329748023725, + "loss": 1.7306, + "step": 14 + }, + { + "epoch": 0.16, + "learning_rate": 0.00019983329399363598, + "loss": 1.7535, + "step": 15 + }, + { + "epoch": 0.17, + "learning_rate": 0.0001997599727063717, + "loss": 1.7811, + "step": 16 + }, + { + "epoch": 0.18, + "learning_rate": 0.000199673343399533, + "loss": 1.7455, + "step": 17 + }, + { + "epoch": 0.19, + "learning_rate": 0.00019957341762950344, + "loss": 1.748, + "step": 18 + }, + { + "epoch": 0.2, + "learning_rate": 0.0001994602087264201, + "loss": 1.723, + "step": 19 + }, + { + "epoch": 0.21, + "learning_rate": 0.00019933373179239502, + "loss": 1.7539, + "step": 20 + }, + { + "epoch": 0.22, + "learning_rate": 0.000199194003699501, + "loss": 1.7293, + "step": 21 + }, + { + "epoch": 0.23, + "learning_rate": 0.0001990410430875205, + "loss": 1.6857, + "step": 22 + }, + { + "epoch": 0.24, + "learning_rate": 0.0001988748703614594, + "loss": 1.7209, + "step": 23 + }, + { + "epoch": 0.25, + "learning_rate": 0.00019869550768882455, + "loss": 1.7127, + "step": 24 + }, + { + "epoch": 0.26, + "learning_rate": 0.0001985029789966671, + "loss": 1.7024, + "step": 25 + }, + { + "epoch": 0.27, + "learning_rate": 0.0001982973099683902, + "loss": 1.6927, + "step": 26 + }, + { + "epoch": 0.28, + "learning_rate": 0.00019807852804032305, + "loss": 1.6584, + "step": 27 + }, + { + "epoch": 0.3, + "learning_rate": 0.0001978466623980609, + "loss": 1.7149, + "step": 28 + }, + { + "epoch": 0.31, + "learning_rate": 0.00019760174397257156, + "loss": 1.6815, + "step": 29 + }, + { + "epoch": 0.32, + "learning_rate": 0.0001973438054360693, + "loss": 1.7067, + "step": 30 + }, + { + "epoch": 0.33, + "learning_rate": 0.00019707288119765623, + "loss": 1.6799, + "step": 31 + }, + { + "epoch": 0.34, + "learning_rate": 0.00019678900739873226, + "loss": 1.7206, + "step": 32 + }, + { + "epoch": 0.35, + "learning_rate": 0.0001964922219081738, + "loss": 1.6458, + "step": 33 + }, + { + "epoch": 0.36, + "learning_rate": 0.00019618256431728194, + "loss": 1.6649, + "step": 34 + }, + { + "epoch": 0.37, + "learning_rate": 0.00019586007593450097, + "loss": 1.6684, + "step": 35 + }, + { + "epoch": 0.38, + "learning_rate": 0.000195524799779908, + "loss": 1.7411, + "step": 36 + }, + { + "epoch": 0.39, + "learning_rate": 0.00019517678057947384, + "loss": 1.6678, + "step": 37 + }, + { + "epoch": 0.4, + "learning_rate": 0.0001948160647590966, + "loss": 1.6738, + "step": 38 + }, + { + "epoch": 0.41, + "learning_rate": 0.00019444270043840852, + "loss": 1.6804, + "step": 39 + }, + { + "epoch": 0.42, + "learning_rate": 0.00019405673742435678, + "loss": 1.6728, + "step": 40 + }, + { + "epoch": 0.43, + "learning_rate": 0.00019365822720455916, + "loss": 1.6941, + "step": 41 + }, + { + "epoch": 0.44, + "learning_rate": 0.00019324722294043558, + "loss": 1.6861, + "step": 42 + }, + { + "epoch": 0.45, + "learning_rate": 0.00019282377946011652, + "loss": 1.6751, + "step": 43 + }, + { + "epoch": 0.46, + "learning_rate": 0.0001923879532511287, + "loss": 1.7175, + "step": 44 + }, + { + "epoch": 0.47, + "learning_rate": 0.00019193980245285966, + "loss": 1.6454, + "step": 45 + }, + { + "epoch": 0.49, + "learning_rate": 0.0001914793868488021, + "loss": 1.6934, + "step": 46 + }, + { + "epoch": 0.5, + "learning_rate": 0.0001910067678585786, + "loss": 1.6534, + "step": 47 + }, + { + "epoch": 0.51, + "learning_rate": 0.00019052200852974819, + "loss": 1.6797, + "step": 48 + }, + { + "epoch": 0.52, + "learning_rate": 0.00019002517352939598, + "loss": 1.6566, + "step": 49 + }, + { + "epoch": 0.53, + "learning_rate": 0.00018951632913550626, + "loss": 1.657, + "step": 50 + }, + { + "epoch": 0.53, + "eval_loss": 1.653587818145752, + "eval_runtime": 200.2473, + "eval_samples_per_second": 6.672, + "eval_steps_per_second": 0.21, + "step": 50 + } + ], + "logging_steps": 1, + "max_steps": 282, + "num_train_epochs": 3, + "save_steps": 50, + "total_flos": 5.1673281053956506e+17, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-50/training_args.bin b/checkpoint-50/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..41fe038f6bd77e41e63090845cfb2375d33e29ec --- /dev/null +++ b/checkpoint-50/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8bdde5c34937c67d70adb2cea431eda05b8cf22eb5ebad425d871a634a8fc84 +size 4411 diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..9bfed7513d3b1b65445af10c4571256f4a19b290 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,6 @@ +{ + "bos_token": "", + "eos_token": "", + "pad_token": "", + "unk_token": "" +} diff --git a/tokenizer.model b/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..77849a9a3ca128f00bc3febdc9b39391a4ef43a3 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,39 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "bos_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "clean_up_tokenization_spaces": false, + "eos_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": null, + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "trust_remote_code": false, + "unk_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "use_default_system_prompt": true, + "use_fast": true +}