Training in progress, step 483

README.md

@@ -3,10 +3,9 @@ library_name: peft
 license: apache-2.0
 base_model: Artples/L-MChat-7b
 tags:
-- axolotl
 - generated_from_trainer
 model-index:
-- name: f356bf41-5032-425b-8da4-284285a6d4b3
+- name: miner_id_24
   results: []
 ---
 
@@ -41,10 +40,10 @@ datasets:
     system_prompt: ''
 debug: null
 deepspeed: null
-early_stopping_patience: 2
+early_stopping_patience: 
 early_stopping_threshold: 1.0e-05
 eval_max_new_tokens: 128
-eval_steps: 138
+eval_steps: 
 eval_strategy: null
 flash_attention: true
 fp16: null
@@ -53,7 +52,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: mrferr3t/f356bf41-5032-425b-8da4-284285a6d4b3
+hub_model_id: 
 hub_repo: null
 hub_strategy: checkpoint
 hub_token: null
@@ -61,7 +60,7 @@ learning_rate: 0.0004
 load_in_4bit: false
 load_in_8bit: false
 local_rank: null
-logging_steps: 138
+logging_steps: 
 lora_alpha: 16
 lora_dropout: 0.05
 lora_fan_in_fan_out: null
@@ -69,7 +68,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: null
+max_steps: 50
 micro_batch_size: 9
 mlflow_experiment_name: /tmp/159279c5560d1ca0_train_data.json
 model_type: AutoModelForCausalLM
@@ -80,7 +79,7 @@ pad_to_sequence_len: true
 resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
-save_steps: 138
+save_steps: 
 saves_per_epoch: 0
 sequence_len: 512
 special_tokens:
@@ -90,7 +89,7 @@ tf32: false
 tokenizer_type: AutoTokenizer
 train_on_inputs: false
 trust_remote_code: true
-val_set_size: .05000000
+val_set_size: 0.0
 wandb_entity: null
 wandb_mode: disabled
 wandb_name: 0da85ac8-626a-4340-8d64-a50f9017e723
@@ -105,11 +104,9 @@ xformers_attention: null
 
 </details><br>
 
-# f356bf41-5032-425b-8da4-284285a6d4b3
+# miner_id_24
 
 This model is a fine-tuned version of [Artples/L-MChat-7b](https://huggingface.co/Artples/L-MChat-7b) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.8026
 
 ## Model description
 
@@ -137,16 +134,10 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 100
+- training_steps: 50
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0015 | 1    | 1.1178          |
-| 3.1374        | 0.2108 | 138  | 0.7750          |
-| 3.0967        | 0.4215 | 276  | 0.7873          |
-| 3.1177        | 0.6323 | 414  | 0.8026          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "o_proj",
+    "v_proj",
+    "gate_proj",
     "up_proj",
     "q_proj",
     "down_proj",
-    "v_proj",
-    "gate_proj"
+    "o_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9990e0e744a6b31e416b1cb206f26ebc6d1304f61fa4266e1ac7e6a9a9a9ee0f
+oid sha256:a270d95a390cb4737825983ffdea47ca6fb7c14f85cfde4db6219c4c0b80bd76
 size 84047370

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc34008e5a29a99fc60ab96d896178d2b4f20a21f213d20b84d7205e7d1e51f4
+oid sha256:825b5fcf901f4a8bf145e82e49be7ae3dc4d2e605a8df0750b726c39d0e0ddd7
 size 83945296

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c4912750e0525668697e67bfbdf8863ec73474c1ab1b2a41b7a08d6ec81c640
+oid sha256:ad2e424f0d3e85ad37a3506914e5792898a4d236d25069f96df22040323b28f0
 size 6776