Instructions to use Gsj49/verl_adaptiveg_checkpoints with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Gsj49/verl_adaptiveg_checkpoints with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Gsj49/verl_adaptiveg_checkpoints", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add checkpoint integrity manifest
Browse files
metadata/items/qwen3-8b-math-common-sketch-step200/checkpoint_manifest.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"created_at": "2026-08-29T18:09:55.259375+00:00",
|
| 3 |
+
"file_count": 14,
|
| 4 |
+
"files": [
|
| 5 |
+
{
|
| 6 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/added_tokens.json",
|
| 7 |
+
"sha256": "c0284b582e14987fbd3d5a2cb2bd139084371ed9acbae488829a1c900833c680",
|
| 8 |
+
"size_bytes": 707
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/chat_template.jinja",
|
| 12 |
+
"sha256": "a55ee1b1660128b7098723e0abcd92caa0788061051c62d51cbe87d9cf1974d8",
|
| 13 |
+
"size_bytes": 4168
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/config.json",
|
| 17 |
+
"sha256": "5987fa06cadf4f720dafad65ea8c401cbb01e7cdce90c68de0a11393d4f0a93f",
|
| 18 |
+
"size_bytes": 1538
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/generation_config.json",
|
| 22 |
+
"sha256": "8b5d33ed01e594c26263de38520e1127731eaaec0e7c35ae92287b274fb6f6ff",
|
| 23 |
+
"size_bytes": 214
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/merges.txt",
|
| 27 |
+
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
|
| 28 |
+
"size_bytes": 1671853
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/model-00001-of-00004.safetensors",
|
| 32 |
+
"sha256": "37a7deb4b5af8851878e04455fd3fd2a138c6d1608639be02389413eebcb4526",
|
| 33 |
+
"size_bytes": 4941139896
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/model-00002-of-00004.safetensors",
|
| 37 |
+
"sha256": "f797c906301c6ad69a41021b4eba882ab65078ee101ea946756eca65938e54f0",
|
| 38 |
+
"size_bytes": 4253268688
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/model-00003-of-00004.safetensors",
|
| 42 |
+
"sha256": "b3253c8a0636d65337410d93e9de2c19622f1f1a61a204a70b85730643887164",
|
| 43 |
+
"size_bytes": 3962679032
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/model-00004-of-00004.safetensors",
|
| 47 |
+
"sha256": "28962ceeccf69b06a94fa42210706640ec670bb3ffbed0d1ad579bd0847eed91",
|
| 48 |
+
"size_bytes": 3224429192
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/model.safetensors.index.json",
|
| 52 |
+
"sha256": "1468b59dbfbfec2221d240fa1a056e29ee204f70ee4f36ec33850aefdd2da0d6",
|
| 53 |
+
"size_bytes": 32914
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/special_tokens_map.json",
|
| 57 |
+
"sha256": "76862e765266b85aa9459767e33cbaf13970f327a0e88d1c65846c2ddd3a1ecd",
|
| 58 |
+
"size_bytes": 613
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/tokenizer.json",
|
| 62 |
+
"sha256": "aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4",
|
| 63 |
+
"size_bytes": 11422654
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/tokenizer_config.json",
|
| 67 |
+
"sha256": "443bfa629eb16387a12edbf92a76f6a6f10b2af3b53d87ba1550adfcf45f7fa0",
|
| 68 |
+
"size_bytes": 5404
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200/vocab.json",
|
| 72 |
+
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
|
| 73 |
+
"size_bytes": 2776833
|
| 74 |
+
}
|
| 75 |
+
],
|
| 76 |
+
"model": {
|
| 77 |
+
"base_model": "Qwen/Qwen3-8B",
|
| 78 |
+
"checkpoint_step": 200,
|
| 79 |
+
"display_name": "Adaptive-G Qwen3-8B Math \u2014 Common Sketch G32/B64",
|
| 80 |
+
"repo_id": "Gsj49/verl_adaptiveg_checkpoints",
|
| 81 |
+
"repo_path": "math_rlvr/qwen3-8b/common_sketch/g32_b64_every25/rep4/global_step_200",
|
| 82 |
+
"slug": "qwen3-8b-math-common-sketch-step200",
|
| 83 |
+
"task": "DAPO-Math-17k RLVR with Math-Verify reward"
|
| 84 |
+
},
|
| 85 |
+
"repo_id": "Gsj49/verl_adaptiveg_checkpoints",
|
| 86 |
+
"size_bytes": 16397433706
|
| 87 |
+
}
|