Text Generation
Transformers
Safetensors
English
phi
Eval Results
Inference Endpoints
text-generation-inference
nano-phi-115M-v0.1 / config.json
kenhktsui's picture
Update config.json
cde2b9c verified
{
"architectures": [
"PhiForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 50256,
"embd_pdrop": 0.0,
"eos_token_id": 50256,
"hidden_act": "gelu_new",
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 1024,
"model_type": "phi",
"num_attention_heads": 24,
"num_hidden_layers": 6,
"num_key_value_heads": 8,
"partial_rotary_factor": 0.4,
"qk_layernorm": false,
"resid_pdrop": 0.1,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.37.2",
"use_cache": true,
"vocab_size": 50295
}