Hugging Face's logo

gpandrad
/
qwen-2.5-3b-r1-countdown

Text Generation
Transformers
TensorBoard
Safetensors
qwen2
Generated from Trainer
trl
grpo
conversational
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
gpandrad commited on
Commit
65dca1c
·
verified ·
1 Parent(s): 021e030

Model save

Browse files
Files changed (4) hide show
  1. all_results.json +4 -4
  2. config.json +1 -1
  3. train_results.json +4 -4
  4. trainer_state.json +14 -8
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 2.9823833263829473e-05,
4
- "train_runtime": 18161.8968,
5
  "train_samples": 45000,
6
- "train_samples_per_second": 1.388,
7
- "train_steps_per_second": 0.025
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 6.298603717921231e-08,
4
+ "train_runtime": 66.8559,
5
  "train_samples": 45000,
6
+ "train_samples_per_second": 376.93,
7
+ "train_steps_per_second": 6.731
8
  }
config.json CHANGED
@@ -23,7 +23,7 @@
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.48.2",
26
- "use_cache": false,
27
  "use_sliding_window": false,
28
  "vocab_size": 151936
29
  }
 
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.48.2",
26
+ "use_cache": true,
27
  "use_sliding_window": false,
28
  "vocab_size": 151936
29
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 2.9823833263829473e-05,
4
- "train_runtime": 18161.8968,
5
  "train_samples": 45000,
6
- "train_samples_per_second": 1.388,
7
- "train_steps_per_second": 0.025
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 6.298603717921231e-08,
4
+ "train_runtime": 66.8559,
5
  "train_samples": 45000,
6
+ "train_samples_per_second": 376.93,
7
+ "train_steps_per_second": 6.731
8
  }
trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.5599626691553896,
5
  "eval_steps": 500,
6
- "global_step": 450,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -2934,13 +2934,19 @@
2934
  "step": 450
2935
  },
2936
  {
2937
- "epoch": 0.5599626691553896,
2938
- "step": 450,
 
 
 
 
 
 
2939
  "total_flos": 0.0,
2940
- "train_loss": 2.9823833263829473e-05,
2941
- "train_runtime": 18161.8968,
2942
- "train_samples_per_second": 1.388,
2943
- "train_steps_per_second": 0.025
2944
  }
2945
  ],
2946
  "logging_steps": 2,
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.5612070306424016,
5
  "eval_steps": 500,
6
+ "global_step": 451,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
2934
  "step": 450
2935
  },
2936
  {
2937
+ "completion_length": 238.5446491241455,
2938
+ "epoch": 0.5612070306424016,
2939
+ "kl": 0.0272674560546875,
2940
+ "reward": 0.9977678656578064,
2941
+ "reward_std": 0.0063134534284472466,
2942
+ "rewards/equation_reward_func": 0.0,
2943
+ "rewards/format_reward_func": 0.9977678656578064,
2944
+ "step": 451,
2945
  "total_flos": 0.0,
2946
+ "train_loss": 6.298603717921231e-08,
2947
+ "train_runtime": 66.8559,
2948
+ "train_samples_per_second": 376.93,
2949
+ "train_steps_per_second": 6.731
2950
  }
2951
  ],
2952
  "logging_steps": 2,