oc06d588da310 / config.json
Jordansky's picture
Upload task output dbe7f337-31b2-489d-a530-17952d63578b
db39cba verified
Raw History Blame Contribute Delete
728 Bytes
{
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 0,
"dtype": "bfloat16",
"eos_token_id": 1,
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 6144,
"max_position_embeddings": 131072,
"mlp_bias": false,
"model_type": "llama",
"num_attention_heads": 16,
"num_hidden_layers": 42,
"num_key_value_heads": 2,
"pad_token_id": 1,
"pretraining_tp": 1,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 5000000,
"rope_type": "default"
},
"tie_word_embeddings": false,
"transformers_version": "5.12.1",
"use_cache": false,
"vocab_size": 130560
}