Model save

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/felipealumni-usp/huggingface/runs/9ucqxz8g)
+[<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/felipealumni-usp/huggingface/runs/6tx4z8at)
 
 This model was trained with SFT.
 

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "o_proj",
+    "q_proj",
+    "gate_proj",
     "down_proj",
     "up_proj",
     "v_proj",
-    "q_proj",
-    "gate_proj"
+    "k_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -3,8 +3,8 @@
     "eval_samples": 26,
     "total_flos": 7.44129634982953e+16,
     "train_loss": 0.0,
-    "train_runtime": 12.907,
+    "train_runtime": 12.3404,
     "train_samples": 25295,
-    "train_samples_per_second": 415.356,
-    "train_steps_per_second": 51.91
+    "train_samples_per_second": 434.426,
+    "train_steps_per_second": 54.293
 }

train_results.json

@@ -3,8 +3,8 @@
     "eval_samples": 26,
     "total_flos": 7.44129634982953e+16,
     "train_loss": 0.0,
-    "train_runtime": 12.907,
+    "train_runtime": 12.3404,
     "train_samples": 25295,
-    "train_samples_per_second": 415.356,
-    "train_steps_per_second": 51.91
+    "train_samples_per_second": 434.426,
+    "train_steps_per_second": 54.293
 }

trainer_state.json

@@ -403,9 +403,9 @@
       "step": 670,
       "total_flos": 7.44129634982953e+16,
       "train_loss": 0.0,
-      "train_runtime": 12.907,
-      "train_samples_per_second": 415.356,
-      "train_steps_per_second": 51.91
+      "train_runtime": 12.3404,
+      "train_samples_per_second": 434.426,
+      "train_steps_per_second": 54.293
     }
   ],
   "logging_steps": 25,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77325f7c9b3efc0699af0bfa2fcc5b4625dbb28ecb6d80c9ebd0d5972803c47f
+oid sha256:61c8a42a344d8a2509af680dba910d2f87a9b97e7400701733aa7fba496865b6
 size 6904