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": [
    0.05917752534151077,
    0.07723169028759003,
    0.09327983111143112,
    0.11935807019472122,
    0.151454359292984,
    0.17953860759735107,
    0.1935807317495346,
    0.20561684668064117,
    0.21163490414619446,
    0.21765294671058655,
    0.22166499495506287,
    0.22166499495506287,
    0.225677028298378,
    0.23370109498500824,
    0.2357071191072464,
    0.24172517657279968,
    0.2457372099161148,
    0.24974924325942993,
    0.25376129150390625,
    0.2517552673816681,
    0.255767285823822,
    0.255767285823822,
    0.2577733099460602,
    0.2577733099460602,
    0.2617853581905365,
    0.26579737663269043,
    0.26980942487716675,
    0.2718154489994049,
    0.27582746744155884,
    0.277833491563797,
    0.3039117157459259,
    0.4463390111923218
  ],
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.36.2",
  "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
}