Model: cygu/llama-2-7b-logit-watermark-distill-kgw-k1-gamma0.25-delta2 Source: Original Platform
26 lines
616 B
JSON
26 lines
616 B
JSON
{
|
|
"_name_or_path": "/scr-ssd/cygu/weights/Llama-2-7b-hf/",
|
|
"architectures": [
|
|
"LlamaForCausalLM"
|
|
],
|
|
"bos_token_id": 1,
|
|
"eos_token_id": 2,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 4096,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 11008,
|
|
"max_position_embeddings": 4096,
|
|
"model_type": "llama",
|
|
"num_attention_heads": 32,
|
|
"num_hidden_layers": 32,
|
|
"num_key_value_heads": 32,
|
|
"pretraining_tp": 1,
|
|
"rms_norm_eps": 1e-05,
|
|
"rope_scaling": null,
|
|
"tie_word_embeddings": false,
|
|
"torch_dtype": "float32",
|
|
"transformers_version": "4.29.2",
|
|
"use_cache": true,
|
|
"vocab_size": 32000
|
|
}
|