Training in progress, step 50

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: number
 tags:
 - generated_from_trainer
-- trl
 - sft
+- trl
 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/3dsmile/huggingface/runs/x4b870bi) 
+[<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/3dsmile/huggingface/runs/0b6v7xuv) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -16,7 +16,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 4,
+  "lora_alpha": 8,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "megatron_config": null,
@@ -25,14 +25,13 @@
   "peft_type": "LORA",
   "peft_version": "0.18.1",
   "qalora_group_size": 16,
-  "r": 4,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "v_proj",
     "q_proj",
-    "up_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 sha256:1be92080d3da3c8370d658e70409ce98d9c469dbf75c4beafa409bcab1256e2c
-size 5308224
+oid sha256:b1d8279786f7f78e61d933bfa2807df8248997d6c9c5b5fddf43f9dc2dcc77f9
+size 5071040

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:053c3229cb7e701a375748667df7194ea6ccf5f8f069762e644901452c0b09e1
+oid sha256:adbc83b8fc704a6a56308798467ddb8b274b083f9cedee2438d5ee3e87635d01
 size 5649