rkazants's picture
Upload 6 files
7a188fb verified
Raw History Blame Contribute Delete
1.4 kB
{
"_orig_attn_implementation": "sdpa",
"architectures": [
"Qwen3DSparkModel"
],
"attention_bias": false,
"attention_dropout": 0.0,
"block_size": 4,
"bos_token_id": 248045,
"draft_vocab_size": 248144,
"dtype": "float32",
"enable_confidence_head": false,
"eos_token_id": [
248044,
248046
],
"head_dim": 32,
"hidden_act": "silu",
"hidden_size": 8,
"initializer_range": 0.02,
"intermediate_size": 16,
"layer_types": [
"full_attention",
"full_attention"
],
"markov_head_type": "vanilla",
"markov_rank": 8,
"mask_token_id": 248076,
"max_position_embeddings": 262144,
"max_window_layers": 28,
"model_type": "qwen3",
"num_attention_heads": 2,
"num_hidden_layers": 2,
"num_key_value_heads": 1,
"num_target_layers": 4,
"pad_token_id": null,
"partial_rotary_factor": 0.25,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"mrope_interleaved": true,
"mrope_section": [
8,
4,
4
],
"partial_rotary_factor": 0.25,
"rope_dim": 32,
"rope_theta": 10000000,
"rope_type": "default"
},
"sample_from_anchor": true,
"sliding_window": null,
"target_layer_ids": [
0,
3
],
"tie_word_embeddings": false,
"transformers_version": "5.10.4",
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 248144
}