Training in progress, step 100

README.md

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: LeL2.1_4B_html1
 tags:
 - generated_from_trainer
+- hf_jobs
 - sft
 - trl
-- hf_jobs
 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/k0ao1tsa) 
+[<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/t2oq44q4) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "down_proj",
-    "v_proj",
     "o_proj",
+    "down_proj",
+    "gate_proj",
     "q_proj",
     "k_proj",
-    "gate_proj"
+    "v_proj",
+    "up_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:d23b97464d113b50311fe53a4960776128ca2c4c6b7b4c5637ed3c35de1d682a
+oid sha256:4cea701ae14c93ac861e3189e1245e7249f5f6e9a7600a0f365f1bce78e4886b
 size 66135336

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:649b92e59027d504a58f9556c88cb62b1ef684df425f27e7850b275a3a77bd57
+oid sha256:9462b0a6f2d30f153928948464ad0dbdeaae7664704c634b1dd7ae3f86c5424c
 size 11422922

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6896d0a33384a23a38793fd95c5c98f231aed35df88b591da5852556c2c4e27e
+oid sha256:cdc4edce8bc1aa9ca4f82902878ec2ab8f4b18321905bf5a3585c02bc57fc126
 size 7121