Safetensors
qwen3
rkazants's picture
Upload 3 files
b9b97e5 verified
Raw
History Blame Contribute Delete
1.03 kB
{
"_orig_attn_implementation": "sdpa",
"architectures": [
"DFlashDraftModel"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": null,
"dflash_config": {
"mask_token_id": 261894,
"target_layer_ids": [
0,
2
]
},
"dtype": "float32",
"eos_token_id": null,
"head_dim": 4,
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 8,
"initializer_range": 0.02,
"intermediate_size": 32,
"layer_types": [
"full_attention",
"full_attention"
],
"max_position_embeddings": 131072,
"max_window_layers": 28,
"model_type": "qwen3",
"num_attention_heads": 2,
"num_hidden_layers": 2,
"num_key_value_heads": 1,
"pad_token_id": null,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 10000.0,
"rope_type": "default"
},
"sliding_window": null,
"tie_word_embeddings": false,
"transformers_version": "5.2.0",
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 262144
}