gemma-7b-orpo / config.json
silviasapora's picture
End of training
c818b7c verified
raw
history blame contribute delete
711 Bytes
{
"_name_or_path": "google/gemma-7b",
"architectures": [
"GemmaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 106,
"eos_token_id": 107,
"head_dim": 256,
"hidden_act": "gelu",
"hidden_activation": "gelu_pytorch_tanh",
"hidden_size": 3072,
"initializer_range": 0.02,
"intermediate_size": 24576,
"max_position_embeddings": 8192,
"model_type": "gemma",
"num_attention_heads": 16,
"num_hidden_layers": 28,
"num_key_value_heads": 16,
"pad_token_id": 107,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000.0,
"torch_dtype": "float32",
"transformers_version": "4.44.2",
"use_cache": true,
"vocab_size": 256000
}