33 lines
758 B
JSON
33 lines
758 B
JSON
|
|
{
|
||
|
|
"architectures": [
|
||
|
|
"GemmaForCausalLM"
|
||
|
|
],
|
||
|
|
"attention_bias": false,
|
||
|
|
"attention_dropout": 0.0,
|
||
|
|
"bos_token_id": 2,
|
||
|
|
"dtype": "bfloat16",
|
||
|
|
"eos_token_id": 1,
|
||
|
|
"head_dim": 64,
|
||
|
|
"hidden_act": "gelu_pytorch_tanh",
|
||
|
|
"hidden_activation": null,
|
||
|
|
"hidden_size": 768,
|
||
|
|
"initializer_range": 0.02,
|
||
|
|
"intermediate_size": 3072,
|
||
|
|
"max_position_embeddings": 512,
|
||
|
|
"model_type": "gemma",
|
||
|
|
"num_attention_heads": 12,
|
||
|
|
"num_hidden_layers": 12,
|
||
|
|
"num_key_value_heads": 1,
|
||
|
|
"pad_token_id": 0,
|
||
|
|
"rms_norm_eps": 1e-06,
|
||
|
|
"rope_parameters": {
|
||
|
|
"rope_theta": 10000.0,
|
||
|
|
"rope_type": "default"
|
||
|
|
},
|
||
|
|
"tie_word_embeddings": true,
|
||
|
|
"transformers_version": "5.0.0",
|
||
|
|
"use_bidirectional_attention": null,
|
||
|
|
"use_cache": false,
|
||
|
|
"vocab_size": 110417
|
||
|
|
}
|