Qwen2.5-Math-7B-NSR / config.json
TianHongZXY's picture
Upload folder using huggingface_hub
8842551 verified
raw
history blame contribute delete
933 Bytes
{
"_name_or_path": "/p/llmresearch/thh9bk/verl-v0.2.0/checkpoints/verl/MATH-Qwen2.5-Math-7B-SFT_negative_advantage-bsz_1024-rollout_n_8-kl_coef_0.0-no_adv_normalization-classic_reward-vllm-0.8.2/global_step_112/actor/huggingface",
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"eos_token_id": 151643,
"hidden_act": "silu",
"hidden_size": 3584,
"initializer_range": 0.02,
"intermediate_size": 18944,
"max_position_embeddings": 4096,
"max_window_layers": 28,
"model_type": "qwen2",
"num_attention_heads": 28,
"num_hidden_layers": 28,
"num_key_value_heads": 4,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000,
"sliding_window": 4096,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.49.0",
"use_cache": true,
"use_mrope": false,
"use_sliding_window": false,
"vocab_size": 152064
}