Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Qwen/Qwen2.5-1.5B-Instruct
|
| 4 |
+
datasets:
|
| 5 |
+
- open-r1/DAPO-Math-17k-Processed
|
| 6 |
+
library_name: transformers
|
| 7 |
+
tags:
|
| 8 |
+
- grpo
|
| 9 |
+
- reinforcement-learning
|
| 10 |
+
- reasoning
|
| 11 |
+
- qwen
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Qwen2.5-1.5B-Instruct_BF16_open-r1-DAPO-Math-17k-Processed_1176_FlashRL_G4-L2048_new
|
| 15 |
+
|
| 16 |
+
This repository contains a checkpoint trained with GRPO on `open-r1/DAPO-Math-17k-Processed` starting from `Qwen/Qwen2.5-1.5B-Instruct`.\
|
| 17 |
+
This snapshot corresponds to training step `1176`.
|
| 18 |
+
|
| 19 |
+
Contents include:
|
| 20 |
+
- Model weights (`.safetensors`)
|
| 21 |
+
- Config files (`config.json`, `generation_config.json`)
|
| 22 |
+
- Tokenizer files (`tokenizer.json`, `tokenizer_config.json`, `vocab.json`, `merges.txt`, `special_tokens_map.json`, `added_tokens.json`)
|
| 23 |
+
- Optional chat template (`chat_template.jinja`)
|
| 24 |
+
|
| 25 |
+
Training artifacts (optimizer/scheduler states and RNG) have been intentionally excluded.
|