config.json

{
  "_name_or_path": "databricks/dbrx-instruct",
  "architectures": [
    "DbrxForCausalLM"
  ],
  "attn_config": {
    "clip_qkv": 8,
    "kv_n_heads": 8,
    "model_type": "",
    "rope_theta": 500000
  },
  "d_model": 6144,
  "emb_pdrop": 0.0,
  "ffn_config": {
    "ffn_hidden_size": 10752,
    "model_type": "",
    "moe_jitter_eps": 0,
    "moe_loss_weight": 0.05,
    "moe_num_experts": 16,
    "moe_top_k": 4
  },
  "initializer_range": 0.02,
  "max_seq_len": 32768,
  "model_type": "dbrx",
  "n_heads": 48,
  "n_layers": 40,
  "output_router_logits": false,
  "quantization_config": {
    "activation_scheme": "static",
    "ignored_layers": [
      "transformer.blocks.26.ffn.router.layer",
      "transformer.blocks.1.ffn.router.layer",
      "transformer.blocks.32.ffn.router.layer",
      "transformer.blocks.35.ffn.router.layer",
      "transformer.blocks.16.ffn.router.layer",
      "transformer.blocks.15.ffn.router.layer",
      "transformer.blocks.10.ffn.router.layer",
      "transformer.blocks.18.ffn.router.layer",
      "transformer.blocks.34.ffn.router.layer",
      "transformer.blocks.38.ffn.router.layer",
      "transformer.blocks.13.ffn.router.layer",
      "transformer.blocks.29.ffn.router.layer",
      "transformer.blocks.3.ffn.router.layer",
      "transformer.blocks.33.ffn.router.layer",
      "transformer.blocks.21.ffn.router.layer",
      "transformer.blocks.19.ffn.router.layer",
      "transformer.blocks.8.ffn.router.layer",
      "transformer.blocks.0.ffn.router.layer",
      "transformer.blocks.37.ffn.router.layer",
      "transformer.blocks.39.ffn.router.layer",
      "transformer.blocks.4.ffn.router.layer",
      "transformer.blocks.28.ffn.router.layer",
      "transformer.blocks.7.ffn.router.layer",
      "transformer.blocks.5.ffn.router.layer",
      "transformer.blocks.9.ffn.router.layer",
      "transformer.blocks.24.ffn.router.layer",
      "transformer.blocks.25.ffn.router.layer",
      "transformer.blocks.20.ffn.router.layer",
      "transformer.blocks.27.ffn.router.layer",
      "transformer.blocks.2.ffn.router.layer",
      "transformer.blocks.31.ffn.router.layer",
      "transformer.blocks.6.ffn.router.layer",
      "transformer.blocks.30.ffn.router.layer",
      "transformer.blocks.36.ffn.router.layer",
      "transformer.blocks.12.ffn.router.layer",
      "transformer.blocks.14.ffn.router.layer",
      "transformer.blocks.23.ffn.router.layer",
      "transformer.blocks.22.ffn.router.layer",
      "transformer.blocks.11.ffn.router.layer",
      "lm_head",
      "transformer.blocks.17.ffn.router.layer"
    ],
    "quant_method": "fp8"
  },
  "resid_pdrop": 0.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.42.1",
  "use_cache": true,
  "vocab_size": 100352
}