emozilla's picture
Upload GPTNeoXLongForCausalLM
e6c7e7e
raw
history blame contribute delete
685 Bytes
{
"_name_or_path": "outputs/pythia-long-6.9b-scifi-fantasy-673-p6144_c1920_y8192",
"architectures": [
"GPTNeoXLongForCausalLM"
],
"bos_token_id": 0,
"eos_token_id": 0,
"hidden_act": "gelu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 16384,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 2048,
"max_positions": 16384,
"model_type": "gpt_neox",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"rotary_emb_base": 10000,
"rotary_pct": 0.25,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.26.1",
"use_cache": false,
"use_parallel_residual": true,
"vocab_size": 50432
}