config.json

{
  "_name_or_path": "meta-llama/Llama-2-7b-hf",
  "architectures": [
    "SparseLlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "ugly_utils.SparseLlamaConfig",
    "AutoModelForCausalLM": "ugly_utils.SparseLlamaForCausalLM"
  },
  "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": "sparse_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,
  "rope_theta": 10000.0,
  "thresholds": [
    0.03109327331185341,
    0.05516548827290535,
    0.07522567361593246,
    0.09327983111143112,
    0.11735205352306366,
    0.13540621101856232,
    0.15546639263629913,
    0.1675025075674057,
    0.17151454091072083,
    0.17552657425403595,
    0.17953860759735107,
    0.18154463171958923,
    0.18154463171958923,
    0.18756268918514252,
    0.18956869840621948,
    0.1935807317495346,
    0.20561684668064117,
    0.2096288800239563,
    0.21364091336727142,
    0.21564693748950958,
    0.22166499495506287,
    0.22367100417613983,
    0.22768303751945496,
    0.22968906164169312,
    0.23169508576393127,
    0.23370109498500824,
    0.23971915245056152,
    0.24172517657279968,
    0.24373118579387665,
    0.24774321913719177,
    0.25376129150390625,
    0.2617853581905365
  ],
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.40.0.dev0",
  "us_sparse_regularization": false,
  "use_cache": false,
  "use_graceful_regularization": false,
  "use_relu": false,
  "use_sparse_model": true,
  "use_sparse_predictor": false,
  "use_sparse_regularization": false,
  "vocab_size": 32000
}