granite-3b-mup / config.json
mayank-mishra's picture
attention multiplier
e1b7bf8
raw
history blame contribute delete
No virus
727 Bytes
{
"attention_bias": true,
"attention_dropout": 0.1,
"attention_multiplier": 0.015625,
"bos_token_id": 0,
"eos_token_id": 0,
"hidden_act": "silu",
"hidden_size": 2304,
"initializer_range": 0.02,
"intermediate_size": 9216,
"embedding_multiplier": 12.0,
"residual_multiplier": 0.22,
"logits_scaling": 9.0,
"max_position_embeddings": 4096,
"mlp_bias": true,
"model_type": "granite",
"num_attention_heads": 36,
"num_hidden_layers": 40,
"num_key_value_heads": 36,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000,
"tie_word_embeddings": true,
"transformers_version": "4.43.0.dev0",
"use_cache": true,
"vocab_size": 49152
}