ssh_1.8B / config.json
Sharathhebbar24's picture
Upload LlamaForCausalLM
007285e verified
{
"_name_or_path": "qnguyen3/quan-1.8b-chat",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": true,
"attention_dropout": 0.0,
"bos_token_id": 151643,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 5504,
"max_position_embeddings": 8192,
"model_type": "llama",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"num_key_value_heads": 16,
"pretraining_tp": 1,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.36.2",
"use_cache": false,
"vocab_size": 151936
}