minerva_1B_v0.1 / config.json
DeepMount00's picture
Upload MistralForCausalLM
a2c5c41 verified
raw
history blame contribute delete
No virus
670 Bytes
{
"_name_or_path": "minerva_boost_2",
"architectures": [
"MistralForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 32768,
"eos_token_id": 32769,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 7168,
"max_position_embeddings": 16384,
"model_type": "mistral",
"num_attention_heads": 16,
"num_hidden_layers": 16,
"num_key_value_heads": 4,
"pad_token_id": 32769,
"rms_norm_eps": 1e-05,
"rope_theta": 10000.0,
"sliding_window": 2048,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.43.0.dev0",
"use_cache": false,
"vocab_size": 32770
}