Training in progress, step 50

README.md

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: LeL2.1_4B_relative2
 tags:
 - generated_from_trainer
+- sft
 - hf_jobs
 - trl
-- sft
 licence: license
 ---
 
@@ -28,7 +28,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/flofiz-universit-de-bourgogne/SFT_Qwen3/runs/b05rb653) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/flofiz-universit-de-bourgogne/SFT_Qwen3/runs/oi6lazf8) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "k_proj",
-    "q_proj",
-    "v_proj",
     "up_proj",
+    "q_proj",
+    "o_proj",
     "down_proj",
-    "o_proj"
+    "k_proj",
+    "v_proj",
+    "gate_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c08f95dc687c653e877c221a284b31c1b02597c802dd1bda9947baa772d23d4
+oid sha256:d667ab7c3454e5fb93e546807606f26f02a3be3bf2b5d4a7d1e1cc0503e3b371
 size 33104520

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b645b8695c6e227c5cc5633712033f24f9659e1ffbd6f6c22026cf2e8282dcc5
+oid sha256:d22efc25463bd38d0d59800c589f0ea290d7f4d6cdd469b523c66dd4a4655b1d
 size 5841