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/bootpin/huggingface/runs/4kxo4msv)
+[<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/bootpin/huggingface/runs/ff380p97)
 
 This model was trained with SFT.
 

all_results.json

@@ -2,8 +2,8 @@
     "epoch": 1.7142857142857144,
     "total_flos": 1113549778714624.0,
     "train_loss": 19.26901610692342,
-    "train_runtime": 147.7595,
+    "train_runtime": 77.6776,
     "train_samples": 460341,
-    "train_samples_per_second": 0.758,
-    "train_steps_per_second": 0.041
+    "train_samples_per_second": 1.442,
+    "train_steps_per_second": 0.077
 }

runs/Apr03_04-51-59_afc374fd6ab1/events.out.tfevents.1743655929.afc374fd6ab1.8698.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4edcb2734e0e6d3eccf420aa51378e9faf3074fd0efd620a151a89ca4abbd36d
+size 7065

train_results.json

@@ -2,8 +2,8 @@
     "epoch": 1.7142857142857144,
     "total_flos": 1113549778714624.0,
     "train_loss": 19.26901610692342,
-    "train_runtime": 147.7595,
+    "train_runtime": 77.6776,
     "train_samples": 460341,
-    "train_samples_per_second": 0.758,
-    "train_steps_per_second": 0.041
+    "train_samples_per_second": 1.442,
+    "train_steps_per_second": 0.077
 }

trainer_state.json

@@ -19,7 +19,7 @@
       "epoch": 0.8571428571428571,
       "eval_loss": 17.570310592651367,
       "eval_runtime": 0.39,
-      "eval_samples_per_second": 5.129,
+      "eval_samples_per_second": 5.128,
       "eval_steps_per_second": 2.564,
       "step": 3
     },
@@ -33,9 +33,9 @@
     {
       "epoch": 1.7142857142857144,
       "eval_loss": 13.667020797729492,
-      "eval_runtime": 0.3996,
-      "eval_samples_per_second": 5.005,
-      "eval_steps_per_second": 2.503,
+      "eval_runtime": 0.3966,
+      "eval_samples_per_second": 5.043,
+      "eval_steps_per_second": 2.522,
       "step": 6
     },
     {
@@ -43,9 +43,9 @@
       "step": 6,
       "total_flos": 1113549778714624.0,
       "train_loss": 19.26901610692342,
-      "train_runtime": 147.7595,
-      "train_samples_per_second": 0.758,
-      "train_steps_per_second": 0.041
+      "train_runtime": 77.6776,
+      "train_samples_per_second": 1.442,
+      "train_steps_per_second": 0.077
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb22ced08cd7f70bdf8f062a4eb6d7546a4d18b7e4a9b91fb45b2b0d376acf5c
+oid sha256:9f22119c867c9141f858801ffc816aee08d01b4f35b0812e21f6d9367d4093e5
 size 7160