diff --git a/Llama-2-13b-chat-hf-inquisitive/README.md b/Llama-2-13b-chat-hf-inquisitive/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-inquisitive/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-inquisitive/adapter_config.json b/Llama-2-13b-chat-hf-inquisitive/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..c298fbbc6d7aa158ab412095f44f8069fb0a3550 --- /dev/null +++ b/Llama-2-13b-chat-hf-inquisitive/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "gate_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "q_proj", + "v_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-inquisitive/adapter_model.bin b/Llama-2-13b-chat-hf-inquisitive/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..ed23aa5944e787d0f34b4f385fc0b853ac39541f --- /dev/null +++ b/Llama-2-13b-chat-hf-inquisitive/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a0fc85ac50b9f042dd01a725597116744c9656f0ee5961ba4af3da2681e91dc +size 1001592141 diff --git a/Llama-2-13b-chat-hf-nudged-questions01/README.md b/Llama-2-13b-chat-hf-nudged-questions01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-nudged-questions01/adapter_config.json b/Llama-2-13b-chat-hf-nudged-questions01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4b3cfab60f28dd7449a46ee2b540700451011 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-nudged-questions01/adapter_model.bin b/Llama-2-13b-chat-hf-nudged-questions01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b7177c28046a60d040613fced0fc6b1eebeab9cb --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71a91be54e7119eff10adf725ba62919c893ffd3dac3167ff98d45498da94e33 +size 1001592141 diff --git a/Llama-2-13b-chat-hf-nudged-questions02/README.md b/Llama-2-13b-chat-hf-nudged-questions02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-nudged-questions02/adapter_config.json b/Llama-2-13b-chat-hf-nudged-questions02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5f7c5abf969d94e3bee6949f0e26e9d357791272 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions02/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "o_proj", + "down_proj", + "q_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-nudged-questions02/adapter_model.bin b/Llama-2-13b-chat-hf-nudged-questions02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..6a573049437edaff1edd97d7724fc23e40a5f33b --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b66fec6bf9330c50747f5a6cedeba4b8641155aea5e6f73f0efc5388c11a622 +size 1001592141 diff --git a/Llama-2-13b-chat-hf-nudged-questions12/README.md b/Llama-2-13b-chat-hf-nudged-questions12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-nudged-questions12/adapter_config.json b/Llama-2-13b-chat-hf-nudged-questions12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4b3cfab60f28dd7449a46ee2b540700451011 --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-nudged-questions12/adapter_model.bin b/Llama-2-13b-chat-hf-nudged-questions12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..75f4e0c1a207c107d132d3aa11f9385aaee19ffe --- /dev/null +++ b/Llama-2-13b-chat-hf-nudged-questions12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8e6c7d86e9adec14775a8562e4a71dd6798bc147b0354ad91274681530fcdf6 +size 1001592141 diff --git a/Llama-2-13b-chat-hf-treccast01/README.md b/Llama-2-13b-chat-hf-treccast01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-treccast01/adapter_config.json b/Llama-2-13b-chat-hf-treccast01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4b3cfab60f28dd7449a46ee2b540700451011 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-treccast01/adapter_model.bin b/Llama-2-13b-chat-hf-treccast01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..e96e78eb2049aa81be0dba4cb9d7835d2394c140 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d69178176e68b4267196175bd2522e1046373803d3f03c76644879fa69066ca +size 1001592141 diff --git a/Llama-2-13b-chat-hf-treccast02/README.md b/Llama-2-13b-chat-hf-treccast02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-treccast02/adapter_config.json b/Llama-2-13b-chat-hf-treccast02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4b3cfab60f28dd7449a46ee2b540700451011 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-treccast02/adapter_model.bin b/Llama-2-13b-chat-hf-treccast02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..723e7e59936a40120d24bfabc58b769675345c3d --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:22a71d55adb121c7956f09af825b681d16206e8b4989d526b854d6dd43a33d7c +size 1001592141 diff --git a/Llama-2-13b-chat-hf-treccast12/README.md b/Llama-2-13b-chat-hf-treccast12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-chat-hf-treccast12/adapter_config.json b/Llama-2-13b-chat-hf-treccast12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4b3cfab60f28dd7449a46ee2b540700451011 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-chat-hf-treccast12/adapter_model.bin b/Llama-2-13b-chat-hf-treccast12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8bc3c342568d5236673a52bedc55cbc499c9de81 --- /dev/null +++ b/Llama-2-13b-chat-hf-treccast12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df395304a13c0d0d5a31e3b376cf582dbb7d1757337e0598d9684a4d9c00e2a1 +size 1001592141 diff --git a/Llama-2-13b-hf-inquisitive/README.md b/Llama-2-13b-hf-inquisitive/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-inquisitive/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-inquisitive/adapter_config.json b/Llama-2-13b-hf-inquisitive/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..232e6c238733056d461f262a0ae73a4f9104f371 --- /dev/null +++ b/Llama-2-13b-hf-inquisitive/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "gate_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "q_proj", + "v_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-inquisitive/adapter_model.bin b/Llama-2-13b-hf-inquisitive/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..a42b81fafddabf68e5176763a8af9b08ed2613e5 --- /dev/null +++ b/Llama-2-13b-hf-inquisitive/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eabe51b49269ca9b6f25deb207b849fdae55018bad4ba6de91834db49b1960c7 +size 1001592141 diff --git a/Llama-2-13b-hf-nudged-questions01/README.md b/Llama-2-13b-hf-nudged-questions01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-nudged-questions01/adapter_config.json b/Llama-2-13b-hf-nudged-questions01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions01/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-nudged-questions01/adapter_model.bin b/Llama-2-13b-hf-nudged-questions01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..5cf2876bc8fc8246599cf7f6f9698602abfd5ac7 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3393b39fed5661c79255eaee759acc554ba47ac9975b344cfc3e299799d51063 +size 1001592141 diff --git a/Llama-2-13b-hf-nudged-questions02/README.md b/Llama-2-13b-hf-nudged-questions02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-nudged-questions02/adapter_config.json b/Llama-2-13b-hf-nudged-questions02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions02/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-nudged-questions02/adapter_model.bin b/Llama-2-13b-hf-nudged-questions02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..5e56752bd1dac81d3303788a7cdad6fe6f0c93ab --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e74a38a83188826aefaec6d3d605442116c0835dba95bf446cac33e576a7100b +size 1001592141 diff --git a/Llama-2-13b-hf-nudged-questions12/README.md b/Llama-2-13b-hf-nudged-questions12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-nudged-questions12/adapter_config.json b/Llama-2-13b-hf-nudged-questions12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions12/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-nudged-questions12/adapter_model.bin b/Llama-2-13b-hf-nudged-questions12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..2480a58767d0f2796810bdf56fc0317e778913e9 --- /dev/null +++ b/Llama-2-13b-hf-nudged-questions12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd014e8d5b07113647e7a39e390c81e9c9d06664825298819784a1e1b650e177 +size 1001592141 diff --git a/Llama-2-13b-hf-treccast01/README.md b/Llama-2-13b-hf-treccast01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-treccast01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-treccast01/adapter_config.json b/Llama-2-13b-hf-treccast01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-treccast01/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-treccast01/adapter_model.bin b/Llama-2-13b-hf-treccast01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..49ee6b09fcceedc89ed95f1be38f6b648b9ec812 --- /dev/null +++ b/Llama-2-13b-hf-treccast01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aece0ab2fe1648018bf30a7847672363cbc233d13168de83efb589f60a3bca6a +size 1001592141 diff --git a/Llama-2-13b-hf-treccast02/README.md b/Llama-2-13b-hf-treccast02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-treccast02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-treccast02/adapter_config.json b/Llama-2-13b-hf-treccast02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-treccast02/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-treccast02/adapter_model.bin b/Llama-2-13b-hf-treccast02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..20e50bdfd6c68803cfdaef0e5618a6134779eaac --- /dev/null +++ b/Llama-2-13b-hf-treccast02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54d15f21ef52fc9e271995b95c7adcbf235b5f6481d4e60b595e9e61aaa69bb9 +size 1001592141 diff --git a/Llama-2-13b-hf-treccast12/README.md b/Llama-2-13b-hf-treccast12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-13b-hf-treccast12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-13b-hf-treccast12/adapter_config.json b/Llama-2-13b-hf-treccast12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a50037c79c072a88dd3de2886b83fe67fa4f4e22 --- /dev/null +++ b/Llama-2-13b-hf-treccast12/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": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-13b-hf-treccast12/adapter_model.bin b/Llama-2-13b-hf-treccast12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..22b8a105cbfa6a104cf6a821f9695d662b932422 --- /dev/null +++ b/Llama-2-13b-hf-treccast12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d7f797d5cd0d54b557b1ab478a700ddd9256dba67e44d27fa649427ca24a140 +size 1001592141 diff --git a/Llama-2-7b-chat-hf-inquisitive/README.md b/Llama-2-7b-chat-hf-inquisitive/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-inquisitive/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-inquisitive/adapter_config.json b/Llama-2-7b-chat-hf-inquisitive/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4d13578bddff1b8994fa20ada78a4d5154fefd87 --- /dev/null +++ b/Llama-2-7b-chat-hf-inquisitive/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "gate_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "q_proj", + "v_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-inquisitive/adapter_model.bin b/Llama-2-7b-chat-hf-inquisitive/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..241c2a6c87201e4ef18b49f8963d77130e487878 --- /dev/null +++ b/Llama-2-7b-chat-hf-inquisitive/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fef1e602f2be97b177dd63c69f73a905c93650bf8121fb4252b283effd0da4dc +size 639792909 diff --git a/Llama-2-7b-chat-hf-nudged-questions01/README.md b/Llama-2-7b-chat-hf-nudged-questions01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-nudged-questions01/adapter_config.json b/Llama-2-7b-chat-hf-nudged-questions01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-nudged-questions01/adapter_model.bin b/Llama-2-7b-chat-hf-nudged-questions01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b85087023e575c5274d3370f45cf042a1b73bfbe --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28b070a71fc829fa8b8a96c4456078093ee4bfde65d4bc4b7e3af00e7444ca18 +size 639792909 diff --git a/Llama-2-7b-chat-hf-nudged-questions02/README.md b/Llama-2-7b-chat-hf-nudged-questions02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-nudged-questions02/adapter_config.json b/Llama-2-7b-chat-hf-nudged-questions02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-nudged-questions02/adapter_model.bin b/Llama-2-7b-chat-hf-nudged-questions02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..42f80aa7b7fcbe7354631153975b1c6b4d630a27 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90f3aaebf72324bf270d951fba355a260ec766cd2735c98cfe1b4d75ecf4646f +size 639792909 diff --git a/Llama-2-7b-chat-hf-nudged-questions12/README.md b/Llama-2-7b-chat-hf-nudged-questions12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-nudged-questions12/adapter_config.json b/Llama-2-7b-chat-hf-nudged-questions12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-nudged-questions12/adapter_model.bin b/Llama-2-7b-chat-hf-nudged-questions12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..eddd462aed51dec5558b12fd04783cb4889c20e3 --- /dev/null +++ b/Llama-2-7b-chat-hf-nudged-questions12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e3fbc6a1a845cc014fe67e3097041a38163cca93030784350eb1e57af91da63 +size 639792909 diff --git a/Llama-2-7b-chat-hf-treccast01/README.md b/Llama-2-7b-chat-hf-treccast01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-treccast01/adapter_config.json b/Llama-2-7b-chat-hf-treccast01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-treccast01/adapter_model.bin b/Llama-2-7b-chat-hf-treccast01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..9b2036f16bc5410cb17a943657b4981cfb59c294 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1dfb766813829e9481c55146c416642035748305f336e2d970a72e860de7c33c +size 639792909 diff --git a/Llama-2-7b-chat-hf-treccast02/README.md b/Llama-2-7b-chat-hf-treccast02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-treccast02/adapter_config.json b/Llama-2-7b-chat-hf-treccast02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-treccast02/adapter_model.bin b/Llama-2-7b-chat-hf-treccast02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..54cfd4c9c5ae99afcbbc36333d3a94a584ec0bdf --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8b09a559084c46ac2642d1a50d4656842dab91c4e28736b9ddf304dee155116 +size 639792909 diff --git a/Llama-2-7b-chat-hf-treccast12/README.md b/Llama-2-7b-chat-hf-treccast12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-chat-hf-treccast12/adapter_config.json b/Llama-2-7b-chat-hf-treccast12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..18cb1b244d8e7a5aeb74295498853bfaad865a40 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-chat-hf-treccast12/adapter_model.bin b/Llama-2-7b-chat-hf-treccast12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..173d697aabd8946d93aaf5365ed1bf95bc70f949 --- /dev/null +++ b/Llama-2-7b-chat-hf-treccast12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:405e0aa1b880758454fafbdd4cdc87a8f369ada3f6874300eed85e3229a9d620 +size 639792909 diff --git a/Llama-2-7b-hf-inquisitive/README.md b/Llama-2-7b-hf-inquisitive/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-inquisitive/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-inquisitive/adapter_config.json b/Llama-2-7b-hf-inquisitive/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0b6e6b0d6afd01b4be712a785bac658fa0b40f2f --- /dev/null +++ b/Llama-2-7b-hf-inquisitive/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "gate_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "q_proj", + "v_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-inquisitive/adapter_model.bin b/Llama-2-7b-hf-inquisitive/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..1e99cb94f023a491a814615a1bb6584f6a36a226 --- /dev/null +++ b/Llama-2-7b-hf-inquisitive/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:563a1802fe110e8758933959d8b46c1561320cd735c8c60493bcc100fa69941c +size 639792909 diff --git a/Llama-2-7b-hf-nudged-questions01/README.md b/Llama-2-7b-hf-nudged-questions01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-nudged-questions01/adapter_config.json b/Llama-2-7b-hf-nudged-questions01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-nudged-questions01/adapter_model.bin b/Llama-2-7b-hf-nudged-questions01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b3fca6d559b90522cd9c67270a72c6e33cca7c9e --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:224b445e61a3649d75fcb8695cd0ea00d9c6f8e9edbcdea1c6621ad4e7be273c +size 639792909 diff --git a/Llama-2-7b-hf-nudged-questions02/README.md b/Llama-2-7b-hf-nudged-questions02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-nudged-questions02/adapter_config.json b/Llama-2-7b-hf-nudged-questions02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-nudged-questions02/adapter_model.bin b/Llama-2-7b-hf-nudged-questions02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..c5793055d4a625f287a399eb5cfe50ed837062f6 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bddf2bfb8e9b02a0527a5d4b08b5538b8e3e3d582d8f956ece27a127cdb4f036 +size 639792909 diff --git a/Llama-2-7b-hf-nudged-questions12/README.md b/Llama-2-7b-hf-nudged-questions12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-nudged-questions12/adapter_config.json b/Llama-2-7b-hf-nudged-questions12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-nudged-questions12/adapter_model.bin b/Llama-2-7b-hf-nudged-questions12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8c79e4ff56d7f4b5c847397e816507128af6f92b --- /dev/null +++ b/Llama-2-7b-hf-nudged-questions12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e50fd5a81ecabb6ecc7b99d270e316cf6cf8d52f765397c447cbfe9ff09c4697 +size 639792909 diff --git a/Llama-2-7b-hf-treccast01/README.md b/Llama-2-7b-hf-treccast01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-treccast01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-treccast01/adapter_config.json b/Llama-2-7b-hf-treccast01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-treccast01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-treccast01/adapter_model.bin b/Llama-2-7b-hf-treccast01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..76f4dc443816feaa9dd260e98545ecd2b9c3973c --- /dev/null +++ b/Llama-2-7b-hf-treccast01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7e86129f26a06a60f8edabf12b57c18c6021dc96c3fe142bb9ef7ce267bf1032 +size 639792909 diff --git a/Llama-2-7b-hf-treccast02/README.md b/Llama-2-7b-hf-treccast02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-treccast02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-treccast02/adapter_config.json b/Llama-2-7b-hf-treccast02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-treccast02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-treccast02/adapter_model.bin b/Llama-2-7b-hf-treccast02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..0e366e7339e713e122203d7efab12a477db7c6b2 --- /dev/null +++ b/Llama-2-7b-hf-treccast02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:986f0b4209a1ba37bde20a9e7b40ca2e6a2757645ed416bf0b6d5403d28c864e +size 639792909 diff --git a/Llama-2-7b-hf-treccast12/README.md b/Llama-2-7b-hf-treccast12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/Llama-2-7b-hf-treccast12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/Llama-2-7b-hf-treccast12/adapter_config.json b/Llama-2-7b-hf-treccast12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..776ed929fca6932913f75276c5f884284a6f5496 --- /dev/null +++ b/Llama-2-7b-hf-treccast12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-7b-hf", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "o_proj", + "down_proj", + "up_proj", + "k_proj", + "v_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/Llama-2-7b-hf-treccast12/adapter_model.bin b/Llama-2-7b-hf-treccast12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..e5e31ea9bc9cbe7731b6106f00971cbae6c00d86 --- /dev/null +++ b/Llama-2-7b-hf-treccast12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c998c92db547f0f0029aca47fb21c1137c3ad254c121c2c602c684b694837de2 +size 639792909 diff --git a/alpaca-7b-nudged-questions01/README.md b/alpaca-7b-nudged-questions01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-nudged-questions01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-nudged-questions01/adapter_config.json b/alpaca-7b-nudged-questions01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-nudged-questions01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-nudged-questions01/adapter_model.bin b/alpaca-7b-nudged-questions01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..dbd0e3035a5e8d58058d05bfa39ce5359881674a --- /dev/null +++ b/alpaca-7b-nudged-questions01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60bb006ecde63542025c9d297d965742d7da7e9f8c05b2643d45266b73f98be5 +size 639792909 diff --git a/alpaca-7b-nudged-questions02/README.md b/alpaca-7b-nudged-questions02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-nudged-questions02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-nudged-questions02/adapter_config.json b/alpaca-7b-nudged-questions02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-nudged-questions02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-nudged-questions02/adapter_model.bin b/alpaca-7b-nudged-questions02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..3140ea7ef26bdd98a5627fe5011fdf33d4ecf8a7 --- /dev/null +++ b/alpaca-7b-nudged-questions02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c84c95043bb04ce002cdd3548980aa252d79085952ab278e21c36a0f89d79e3 +size 639792909 diff --git a/alpaca-7b-nudged-questions12/README.md b/alpaca-7b-nudged-questions12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-nudged-questions12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-nudged-questions12/adapter_config.json b/alpaca-7b-nudged-questions12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-nudged-questions12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-nudged-questions12/adapter_model.bin b/alpaca-7b-nudged-questions12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b389e496b37b7bac67fe9a9762c9f3f3d1e411c6 --- /dev/null +++ b/alpaca-7b-nudged-questions12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e4940725529546fb6a555e9315dbe19897ea5a72fa8367e18a6faa05093a3a6 +size 639792909 diff --git a/alpaca-7b-treccast01/README.md b/alpaca-7b-treccast01/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-treccast01/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-treccast01/adapter_config.json b/alpaca-7b-treccast01/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-treccast01/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-treccast01/adapter_model.bin b/alpaca-7b-treccast01/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..d3966352fb5c4e2f5da3cfba08455f7a087c4dc9 --- /dev/null +++ b/alpaca-7b-treccast01/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7824fe7f0815a22924ddcbea9f303430da202ebcf3adc04ec74b730185d5e728 +size 639792909 diff --git a/alpaca-7b-treccast02/README.md b/alpaca-7b-treccast02/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-treccast02/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-treccast02/adapter_config.json b/alpaca-7b-treccast02/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-treccast02/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-treccast02/adapter_model.bin b/alpaca-7b-treccast02/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..eea56b3d0b7250e18bb9db78384d63f6eae07120 --- /dev/null +++ b/alpaca-7b-treccast02/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5850d5fb87eb2f0a8815e5717a33f9b337ae5ae810eec928eb11316d174e4416 +size 639792909 diff --git a/alpaca-7b-treccast12/README.md b/alpaca-7b-treccast12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30669553d106890c3a6b73a44938474314a55ce8 --- /dev/null +++ b/alpaca-7b-treccast12/README.md @@ -0,0 +1,20 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- 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: False +- bnb_4bit_compute_dtype: float16 +### Framework versions + + +- PEFT 0.4.0 diff --git a/alpaca-7b-treccast12/adapter_config.json b/alpaca-7b-treccast12/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6918db8e4207cfe9fbee7bb60657a217a2ee2dcf --- /dev/null +++ b/alpaca-7b-treccast12/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "./alpaca/alpaca-7b/", + "bias": "none", + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.1, + "modules_to_save": null, + "peft_type": "LORA", + "r": 64, + "revision": null, + "target_modules": [ + "q_proj", + "gate_proj", + "k_proj", + "up_proj", + "v_proj", + "o_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/alpaca-7b-treccast12/adapter_model.bin b/alpaca-7b-treccast12/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..ff8fcd0c5a40728af1895a9a71f188c6a4ee44bb --- /dev/null +++ b/alpaca-7b-treccast12/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e5e0e67dd62c04b0ce7a8757ce0e20f509ce4f1513d68c3a24524dcde99b5f52 +size 639792909