Training in progress, epoch 1

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: kto_simpletext
 tags:
 - generated_from_trainer
-- trl
 - unsloth
+- trl
 - kto
 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/ioakeime-aristotle-university-of-thessaloniki/kto_simpletext/runs/09vukwft) 
+[<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/ioakeime-aristotle-university-of-thessaloniki/kto_simpletext/runs/f39llvr3) 
 
 
 This model was trained with KTO, a method introduced in [KTO: Model Alignment as Prospect Theoretic Optimization](https://huggingface.co/papers/2402.01306).

adapter_config.json

@@ -35,12 +35,12 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "up_proj",
-    "down_proj",
     "gate_proj",
+    "o_proj",
     "v_proj",
     "q_proj",
-    "o_proj"
+    "down_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 sha256:98c8db71a4e798db3370d3def3b4a8d110e1e4e25adc55c1f3b2f7397dca4780
+oid sha256:a21dcb8a40c4bfec72369a7fef5614c254fa25c129f5fb9d7ccc23ea440899c5
 size 167832240

tokenizer_config.json

@@ -6179,7 +6179,7 @@
   "legacy": false,
   "model_max_length": 32768,
   "pad_token": "[control_768]",
-  "padding_side": "left",
+  "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffd609922f46985ec8c10ea1892a07c8c63fa155b2c83ec6556432cd32f916ec
+oid sha256:fd35f28e7f19d247eefdbc466c9d46498bc2fa6d1debd8940b021421e74a451c
 size 6481