Training in progress, step 1000

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: gemma-3-4b-4q-roman-it
 tags:
 - generated_from_trainer
-- sft
 - trl
+- sft
 licence: license
 ---
 
@@ -27,7 +27,7 @@ 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/amirdanger80p-commit-m/huggingface/runs/8z6r7e43) 
+[<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/amirdanger80p-commit-m/huggingface/runs/e5jdka94) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
     "up_proj",
+    "gate_proj",
     "v_proj",
     "k_proj",
+    "down_proj",
     "q_proj",
-    "gate_proj",
-    "down_proj"
+    "o_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:d3b686ae1fc4d3e7ab9c7ecedc2908d703542f07f0adefb6d6bfc4cdcdf41eeb
+oid sha256:78edd62ac14f167fc94c8ec2a5753073e93a392cf3e905ad721175dbc9ccfb89
 size 131252288

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75487669ad5a5004dca3c5049f412dbace206c680f2ca4d40f320e04c8d2bdd7
+oid sha256:c88868d7889df046a30964fea7e33f8e96c9f674445c8f5178d4e396853acdeb
 size 6161