RE_Model_Copy / config.json
C1pher's picture
Upload folder using huggingface_hub
c6310e7 verified
raw
history blame contribute delete
924 Bytes
{
"architectures": [
"Qwen2MergeForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 151646,
"eos_token_id": 151643,
"hidden_act": "silu",
"hidden_size": 1536,
"initializer_range": 0.02,
"intermediate_size": 8960,
"max_position_embeddings": 131072,
"max_window_layers": 21,
"model_type": "qwen2_merge",
"num_attention_heads": 12,
"num_hidden_layers": 29,
"num_key_value_heads": 2,
"pad_token_id": 151643,
"predictor_num_hidden_layers": 1,
"rescale_layer_end": 28,
"rescale_layer_start": 0,
"rescale_lora_rank": 32,
"rescale_magnitude_max": 1.5,
"rescale_magnitude_min": -0.5,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.51.3",
"use_cache": false,
"use_mrope": false,
"use_sliding_window": false,
"vocab_size": 151936
}