End of training

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/paolordls-upd/Crosslingual%20Lens/runs/4kqkipyh)
+[<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/paolordls-upd/Crosslingual%20Lens/runs/ck16vjd1)
 
 This model was trained with SFT.
 

adapter_config.json

@@ -10,8 +10,8 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
-  "lora_dropout": 0.05,
+  "lora_alpha": 64,
+  "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
@@ -20,9 +20,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "gate_proj",
     "k_proj",
-    "v_proj",
-    "q_proj"
+    "up_proj",
+    "q_proj",
+    "down_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a8ebe8f4b8a71ebed3a9ac42f4fd2cf8fc77577a83f8f35710a9701abed0a9c
-size 9179192
+oid sha256:ca48d1ca822cc2c23edd75e18a5b61085b7d746c5c8fbd8e849327858228d160
+size 37953560