78 lines
1.7 KiB
JSON
78 lines
1.7 KiB
JSON
{
|
|
"architectures": [
|
|
"Qwen3ForCausalLM"
|
|
],
|
|
"attention_bias": false,
|
|
"attention_dropout": 0.0,
|
|
"bos_token_id": 50030,
|
|
"dtype": "bfloat16",
|
|
"eos_token_id": 50031,
|
|
"head_dim": 96,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 1152,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 2880,
|
|
"layer_types": [
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention",
|
|
"full_attention"
|
|
],
|
|
"max_position_embeddings": 8192,
|
|
"max_window_layers": 40,
|
|
"model_name": "test_checkpoint",
|
|
"model_type": "qwen3",
|
|
"num_attention_heads": 8,
|
|
"num_hidden_layers": 40,
|
|
"num_key_value_heads": 8,
|
|
"pad_token_id": 50034,
|
|
"rms_norm_eps": 1e-06,
|
|
"rope_parameters": {
|
|
"rope_theta": 1000000,
|
|
"rope_type": "default"
|
|
},
|
|
"sliding_window": null,
|
|
"tie_word_embeddings": false,
|
|
"transformers_version": "5.3.0",
|
|
"unsloth_version": "2026.3.4",
|
|
"use_cache": false,
|
|
"use_sliding_window": false,
|
|
"vocab_size": 50048
|
|
}
|