File size: 1,318 Bytes
b71f8a6 8841574 b71f8a6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
{
"architectures": [
"SharedSpaceDecoderForCausalLM"
],
"auto_map": {
"AutoConfig": "configuration_shared_subspace_decoder.SharedSpaceDecoderConfig",
"AutoModel": "modeling_shared_subspace_decoder.SharedSpaceDecoderModel",
"AutoModelForCausalLM": "modeling_shared_subspace_decoder.SharedSpaceDecoderForCausalLM"
},
"attention_backend": "flash_attention_2",
"attention_bias": false,
"attention_dropout_prob": 0.1,
"bos_token_id": 50256,
"classifier_dropout": null,
"dtype": "float32",
"eos_token_id": 50256,
"ffn_decompose": false,
"ffn_rank": null,
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 2048,
"kv_shared_dim": 96,
"layer_norm_eps": 1e-12,
"max_position_embeddings": 1024,
"model_type": "shared_subspace_decoder",
"nope_dims": 32,
"norm_type": "rmsnorm",
"num_attention_heads": 12,
"num_dense_layers": 0,
"num_hidden_layers": 12,
"o_shared_dim": null,
"pad_token_id": 50256,
"q_shared_dim": 192,
"qk_private_dim": 64,
"rms_norm_eps": 1e-06,
"rope_dims": 32,
"rope_scaling": {
"factor": 2.0,
"type": "linear"
},
"rope_theta": 10000.0,
"transformers_version": "4.56.0",
"vo_private_dim": 64,
"vocab_rank": null,
"vocab_size": 50257,
"vocab_subspace": false
}
|