Instructions to use NTQuoc/OpenRS-GRPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NTQuoc/OpenRS-GRPO with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("NTQuoc/OpenRS-GRPO", dtype="auto") - Notebooks
- Google Colab
- Kaggle
End of training
Browse files- README.md +3 -1
- config.json +52 -42
README.md
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
---
|
| 2 |
base_model: Qwen/Qwen3.5-0.8B
|
|
|
|
| 3 |
library_name: transformers
|
| 4 |
model_name: OpenRS-GRPO
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
|
|
|
| 7 |
- trl
|
| 8 |
- grpo
|
| 9 |
licence: license
|
|
@@ -11,7 +13,7 @@ licence: license
|
|
| 11 |
|
| 12 |
# Model Card for OpenRS-GRPO
|
| 13 |
|
| 14 |
-
This model is a fine-tuned version of [Qwen/Qwen3.5-0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B).
|
| 15 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
|
| 17 |
## Quick start
|
|
|
|
| 1 |
---
|
| 2 |
base_model: Qwen/Qwen3.5-0.8B
|
| 3 |
+
datasets: knoveleng/open-rs
|
| 4 |
library_name: transformers
|
| 5 |
model_name: OpenRS-GRPO
|
| 6 |
tags:
|
| 7 |
- generated_from_trainer
|
| 8 |
+
- open-r1
|
| 9 |
- trl
|
| 10 |
- grpo
|
| 11 |
licence: license
|
|
|
|
| 13 |
|
| 14 |
# Model Card for OpenRS-GRPO
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [Qwen/Qwen3.5-0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B) on the [knoveleng/open-rs](https://huggingface.co/datasets/knoveleng/open-rs) dataset.
|
| 17 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 18 |
|
| 19 |
## Quick start
|
config.json
CHANGED
|
@@ -1,62 +1,72 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"Qwen2ForCausalLM"
|
| 4 |
-
],
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
|
|
|
|
|
|
|
|
|
| 9 |
"hidden_act": "silu",
|
| 10 |
-
"hidden_size":
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
-
"intermediate_size":
|
| 13 |
"layer_types": [
|
|
|
|
|
|
|
|
|
|
| 14 |
"full_attention",
|
|
|
|
|
|
|
|
|
|
| 15 |
"full_attention",
|
|
|
|
|
|
|
|
|
|
| 16 |
"full_attention",
|
|
|
|
|
|
|
|
|
|
| 17 |
"full_attention",
|
|
|
|
|
|
|
|
|
|
| 18 |
"full_attention",
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"full_attention",
|
| 23 |
-
"full_attention",
|
| 24 |
-
"full_attention",
|
| 25 |
-
"full_attention",
|
| 26 |
-
"full_attention",
|
| 27 |
-
"full_attention",
|
| 28 |
-
"full_attention",
|
| 29 |
-
"full_attention",
|
| 30 |
-
"full_attention",
|
| 31 |
-
"full_attention",
|
| 32 |
-
"full_attention",
|
| 33 |
-
"full_attention",
|
| 34 |
-
"full_attention",
|
| 35 |
-
"full_attention",
|
| 36 |
-
"full_attention",
|
| 37 |
-
"full_attention",
|
| 38 |
-
"full_attention",
|
| 39 |
-
"full_attention",
|
| 40 |
-
"full_attention",
|
| 41 |
"full_attention"
|
| 42 |
],
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"
|
| 46 |
-
"
|
| 47 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
"num_key_value_heads": 2,
|
| 49 |
-
"pad_token_id":
|
|
|
|
| 50 |
"rms_norm_eps": 1e-06,
|
| 51 |
"rope_parameters": {
|
| 52 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
"rope_type": "default"
|
| 54 |
},
|
| 55 |
-
"
|
| 56 |
-
"tie_word_embeddings": false,
|
| 57 |
"transformers_version": "5.8.0.dev0",
|
| 58 |
"use_cache": true,
|
| 59 |
-
"
|
| 60 |
-
"use_sliding_window": false,
|
| 61 |
-
"vocab_size": 151936
|
| 62 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"attention_bias": false,
|
|
|
|
|
|
|
| 3 |
"attention_dropout": 0.0,
|
| 4 |
+
"attn_output_gate": true,
|
| 5 |
+
"bos_token_id": null,
|
| 6 |
+
"dtype": "bfloat16",
|
| 7 |
+
"eos_token_id": 248046,
|
| 8 |
+
"full_attention_interval": 4,
|
| 9 |
+
"head_dim": 256,
|
| 10 |
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3584,
|
| 14 |
"layer_types": [
|
| 15 |
+
"linear_attention",
|
| 16 |
+
"linear_attention",
|
| 17 |
+
"linear_attention",
|
| 18 |
"full_attention",
|
| 19 |
+
"linear_attention",
|
| 20 |
+
"linear_attention",
|
| 21 |
+
"linear_attention",
|
| 22 |
"full_attention",
|
| 23 |
+
"linear_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"linear_attention",
|
| 26 |
"full_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
"full_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
"full_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"linear_attention",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
"full_attention"
|
| 39 |
],
|
| 40 |
+
"linear_conv_kernel_dim": 4,
|
| 41 |
+
"linear_key_head_dim": 128,
|
| 42 |
+
"linear_num_key_heads": 16,
|
| 43 |
+
"linear_num_value_heads": 16,
|
| 44 |
+
"linear_value_head_dim": 128,
|
| 45 |
+
"mamba_ssm_dtype": "float32",
|
| 46 |
+
"max_position_embeddings": 262144,
|
| 47 |
+
"mlp_only_layers": [],
|
| 48 |
+
"model_type": "qwen3_5_text",
|
| 49 |
+
"mtp_num_hidden_layers": 1,
|
| 50 |
+
"mtp_use_dedicated_embeddings": false,
|
| 51 |
+
"num_attention_heads": 8,
|
| 52 |
+
"num_hidden_layers": 24,
|
| 53 |
"num_key_value_heads": 2,
|
| 54 |
+
"pad_token_id": 248044,
|
| 55 |
+
"partial_rotary_factor": 0.25,
|
| 56 |
"rms_norm_eps": 1e-06,
|
| 57 |
"rope_parameters": {
|
| 58 |
+
"mrope_interleaved": true,
|
| 59 |
+
"mrope_section": [
|
| 60 |
+
11,
|
| 61 |
+
11,
|
| 62 |
+
10
|
| 63 |
+
],
|
| 64 |
+
"partial_rotary_factor": 0.25,
|
| 65 |
+
"rope_theta": 10000000,
|
| 66 |
"rope_type": "default"
|
| 67 |
},
|
| 68 |
+
"tie_word_embeddings": true,
|
|
|
|
| 69 |
"transformers_version": "5.8.0.dev0",
|
| 70 |
"use_cache": true,
|
| 71 |
+
"vocab_size": 248320
|
|
|
|
|
|
|
| 72 |
}
|