Model save

README.md

@@ -26,7 +26,7 @@ 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/757885345/huggingface/runs/j2m7tvu0) 
+[<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/757885345/huggingface/runs/hqhsm4kq) 
 
 
 This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
@@ -34,7 +34,7 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
 ### Framework versions
 
 - TRL: 0.14.0
-- Transformers: 4.48.2
+- Transformers: 4.49.0.dev0
 - Pytorch: 2.5.1
 - Datasets: 3.2.0
 - Tokenizers: 0.21.0

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
     "train_loss": 1.862645149230957e-09,
-    "train_runtime": 21.2071,
+    "train_runtime": 10.39,
     "train_samples": 6,
-    "train_samples_per_second": 0.283,
-    "train_steps_per_second": 0.047
+    "train_samples_per_second": 0.577,
+    "train_steps_per_second": 0.096
 }

config.json

@@ -19,10 +19,10 @@
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
   "rope_theta": 1000000.0,
-  "sliding_window": null,
+  "sliding_window": 32768,
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.48.2",
+  "transformers_version": "4.49.0.dev0",
   "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936

generation_config.json

@@ -10,5 +10,5 @@
   "temperature": 0.7,
   "top_k": 20,
   "top_p": 0.8,
-  "transformers_version": "4.48.2"
+  "transformers_version": "4.49.0.dev0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:130282af0dfa9fe5840737cc49a0d339d06075f83c5a315c3372c9a0740d0b96
-size 988097824
+oid sha256:8f301d59ea1204790a259d51cce69fe450fc2ebb8d3f7b906f8c4a2dd5c787d8
+size 1260367448

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
     "train_loss": 1.862645149230957e-09,
-    "train_runtime": 21.2071,
+    "train_runtime": 10.39,
     "train_samples": 6,
-    "train_samples_per_second": 0.283,
-    "train_steps_per_second": 0.047
+    "train_samples_per_second": 0.577,
+    "train_steps_per_second": 0.096
 }

trainer_state.json

@@ -21,9 +21,9 @@
       "step": 1,
       "total_flos": 0.0,
       "train_loss": 1.862645149230957e-09,
-      "train_runtime": 21.2071,
-      "train_samples_per_second": 0.283,
-      "train_steps_per_second": 0.047
+      "train_runtime": 10.39,
+      "train_samples_per_second": 0.577,
+      "train_steps_per_second": 0.096
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5790cfe1b5120af98ada14eba1bd93b2e2a008dd6cdd1bbde119697a4f17791b
+oid sha256:acd50cb540127364ff247e65aed68c58de505ca2d10bceacd7e53b909ffc3d10
 size 7288