File size: 1,277 Bytes
a47a33d 8265e0f a47a33d f4ca790 8681515 3cc5dd0 a47a33d 3cc5dd0 a47a33d cde9322 8265e0f a47a33d cde9322 a47a33d 8681515 a47a33d cde9322 a47a33d 8681515 cde9322 a47a33d 3cc5dd0 |
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 |
{
"architectures": [
"ArcticForCausalLM"
],
"auto_map": {
"AutoConfig": "configuration_arctic.ArcticConfig",
"AutoModel": "modeling_arctic.ArcticModel",
"AutoModelForCausalLM": "modeling_arctic.ArcticForCausalLM",
"AutoModelForSequenceClassification": "modeling_arctic.ArcticForSequenceClassification"
},
"attention_dropout": 0,
"bos_token_id": 31998,
"enable_expert_tensor_parallelism": false,
"eos_token_id": 31999,
"hidden_act": "silu",
"hidden_size": 7168,
"initializer_range": 0.02,
"intermediate_size": 4864,
"max_position_embeddings": 4096,
"max_sequence_length": 4096,
"model_type": "arctic",
"moe_eval_capacity_factor": 1,
"moe_layer_frequency": 1,
"moe_min_capacity": 0,
"moe_token_dropping": false,
"moe_train_capacity_factor": 1,
"num_attention_heads": 56,
"num_experts_per_tok": 2,
"num_hidden_layers": 35,
"num_key_value_heads": 8,
"num_local_experts": 128,
"parallel_attn_mlp_res": true,
"quantization": null,
"rms_norm_eps": 1e-05,
"rope_theta": 10000,
"router_aux_loss_coef": 0.001,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.0.dev0",
"use_cache": true,
"use_residual": true,
"vocab_size": 32000
}
|