Training in progress, step 1793

README.md

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: m-lora-3
 tags:
 - generated_from_trainer
-- trl
 - unsloth
 - sft
+- trl
 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/batuhan409/huggingface/runs/btmz7cx0) 
+[<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/batuhan409/huggingface/runs/4qg2fop8) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -22,28 +22,25 @@
   "lora_dropout": 0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
-  "modules_to_save": [
-    "embed_tokens",
-    "lm_head"
-  ],
+  "modules_to_save": null,
   "peft_type": "LORA",
   "qalora_group_size": 16,
-  "r": 32,
+  "r": 128,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
+    "down_proj",
     "q_proj",
-    "gate_proj",
-    "o_proj",
     "k_proj",
-    "down_proj",
-    "up_proj",
-    "v_proj"
+    "o_proj",
+    "v_proj",
+    "gate_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,
-  "use_dora": true,
+  "use_dora": false,
   "use_qalora": false,
-  "use_rslora": false
+  "use_rslora": true
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb9178bcb130570f3c56276872aa0cf4b5263ecaeeb698f9be3af1a18b2c211e
-size 832034064
+oid sha256:4d25a547ed4d10b0cc6a3a698f830e443633a68cf58b5bdb730d00501bbdb83b
+size 218134048

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aafdac56c2e106724ac5bea1f4e43a4d8db414f442eface15b2bf7c4461ccf48
+oid sha256:5e2c0fb7d887bf63501113ad3e27c7a29b06c831e81b3dc8ffceaf0731863714
 size 6225