Training in progress, step 4

README.md

@@ -1,6 +1,5 @@
 ---
 base_model: openai/gpt-oss-20b
-datasets: Trelis/Multilingual-Thinking
 library_name: transformers
 model_name: gpt-oss-20b-multilingual-reasoner
 tags:
@@ -12,7 +11,7 @@ licence: license
 
 # Model Card for gpt-oss-20b-multilingual-reasoner
 
-This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) on the [Trelis/Multilingual-Thinking](https://huggingface.co/datasets/Trelis/Multilingual-Thinking) dataset.
+This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

adapter_config.json

@@ -29,17 +29,13 @@
   "revision": null,
   "target_modules": [
     "v_proj",
-    "k_proj",
     "q_proj",
+    "k_proj",
     "o_proj"
   ],
   "target_parameters": [
-    "7.mlp.experts.gate_up_proj",
-    "7.mlp.experts.down_proj",
-    "15.mlp.experts.gate_up_proj",
-    "15.mlp.experts.down_proj",
-    "23.mlp.experts.gate_up_proj",
-    "23.mlp.experts.down_proj"
+    "down_proj",
+    "up_proj"
   ],
   "task_type": null,
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e571fceb4d0c0bf8def59f117362334250df32a016ab0d528a2ccba11cf4af65
-size 60189176
+oid sha256:34e7e71856637583becf65c60619a073422601a908b9ff28cc74e7548450df28
+size 157515072

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d00d2653732d4483cdff9efa7615137cc0a95de867946029bfeffffa416fa47
-size 6225
+oid sha256:095593e6b0aabba7b77c928fc41f28b6613b2ab4971641382afa29c821df33ae
+size 6161