Training in progress, step 300

README.md

@@ -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/dvmegrnl) 
+[<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/x0bsg7bv) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -15,7 +15,7 @@
   "loftq_config": {},
   "lora_alpha": 64,
   "lora_bias": false,
-  "lora_dropout": 0.0,
+  "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
@@ -25,12 +25,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "qkv",
+    "v_proj",
+    "k_proj",
     "o_proj",
     "q_proj",
-    "proj",
-    "v_proj",
-    "k_proj"
+    "qkv",
+    "proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22e03d40aae20fe91af7dcbdbf2ce089c543812b7af83fb586067ec8fff6ba39
+oid sha256:23f424e052b084d1d15dede781ce3e672ecbdac5d61869887ec8cf8b74b051ea
 size 181568928

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b40f1721f2a7c54e0ced2f9cefdc6dadf3ec54e407401f6efc56c7501aca229e
+oid sha256:41e04dec62d411c593147e4a074736111babbd5b975fa09d849ef8f8331ed939
 size 5777