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/deepakkarkala-personal/huggingface/runs/irpr7u7w) 
+[<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/deepakkarkala-personal/huggingface/runs/l8vng39l) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -24,12 +24,12 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "gate_proj",
     "up_proj",
+    "down_proj",
     "q_proj",
     "o_proj",
-    "down_proj",
-    "v_proj"
+    "v_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 2381129238708224.0,
-    "train_loss": 1.763643136391273,
-    "train_runtime": 450.6133,
+    "train_loss": 0.0,
+    "train_runtime": 0.6875,
     "train_samples": 2079,
-    "train_samples_per_second": 4.614,
-    "train_steps_per_second": 0.577
+    "train_samples_per_second": 3024.065,
+    "train_steps_per_second": 378.19
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 2381129238708224.0,
-    "train_loss": 1.763643136391273,
-    "train_runtime": 450.6133,
+    "train_loss": 0.0,
+    "train_runtime": 0.6875,
     "train_samples": 2079,
-    "train_samples_per_second": 4.614,
-    "train_steps_per_second": 0.577
+    "train_samples_per_second": 3024.065,
+    "train_steps_per_second": 378.19
 }

trainer_state.json

@@ -432,23 +432,14 @@
       "mean_token_accuracy": 0.5999625205993653,
       "step": 260
     },
-    {
-      "epoch": 1.0,
-      "eval_loss": 1.6983668804168701,
-      "eval_mean_token_accuracy": 0.6045786820608994,
-      "eval_runtime": 14.6795,
-      "eval_samples_per_second": 15.804,
-      "eval_steps_per_second": 1.976,
-      "step": 260
-    },
     {
       "epoch": 1.0,
       "step": 260,
       "total_flos": 2381129238708224.0,
-      "train_loss": 1.763643136391273,
-      "train_runtime": 450.6133,
-      "train_samples_per_second": 4.614,
-      "train_steps_per_second": 0.577
+      "train_loss": 0.0,
+      "train_runtime": 0.6875,
+      "train_samples_per_second": 3024.065,
+      "train_steps_per_second": 378.19
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb2f4a9068f2c81f9c25697790afcbc1b63d822cc8baf2b32e460d4be9517e55
+oid sha256:79d51c1e8491754d2fbaafbfc2fc1643064cc3c3e48805bb3ed7628185435573
 size 5688