35 lines
832 B
JSON
35 lines
832 B
JSON
{
|
|
"_name_or_path": "jphme/em_german_13b_v01",
|
|
"architectures": [
|
|
"LlamaForCausalLM"
|
|
],
|
|
"bos_token_id": 1,
|
|
"eos_token_id": 2,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 5120,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 13824,
|
|
"max_position_embeddings": 4096,
|
|
"model_type": "llama",
|
|
"num_attention_heads": 40,
|
|
"num_hidden_layers": 40,
|
|
"num_key_value_heads": 40,
|
|
"quantization_config": {
|
|
"bits": 4,
|
|
"damp_percent": 0.1,
|
|
"desc_act": 1,
|
|
"group_size": 32,
|
|
"quant_method": "gptq",
|
|
"sym": true,
|
|
"true_sequential": true,
|
|
"model_file_base_name": "model"
|
|
},
|
|
"rms_norm_eps": 1e-05,
|
|
"rope_scaling": null,
|
|
"rope_theta": 10000.0,
|
|
"tie_word_embeddings": false,
|
|
"torch_dtype": "float16",
|
|
"transformers_version": "4.34.0.dev0",
|
|
"use_cache": true,
|
|
"vocab_size": 32000
|
|
} |