config.json

{
  "vocab_size": 128256,
  "max_position_embeddings": 8192,
  "hidden_size": 4096,
  "intermediate_size": 14336,
  "num_hidden_layers": 32,
  "num_attention_heads": 32,
  "num_key_value_heads": 8,
  "hidden_act": "silu",
  "initializer_range": 0.02,
  "rms_norm_eps": 1e-05,
  "pretraining_tp": 1,
  "use_cache": true,
  "rope_theta": 500000.0,
  "rope_scaling": null,
  "attention_bias": false,
  "attention_dropout": 0.0,
  "return_dict": true,
  "output_hidden_states": false,
  "output_attentions": false,
  "torchscript": false,
  "torch_dtype": "float16",
  "use_bfloat16": false,
  "tf_legacy_loss": false,
  "pruned_heads": {},
  "tie_word_embeddings": false,
  "chunk_size_feed_forward": 0,
  "is_encoder_decoder": false,
  "is_decoder": false,
  "cross_attention_hidden_size": null,
  "add_cross_attention": false,
  "tie_encoder_decoder": false,
  "max_length": 20,
  "min_length": 0,
  "do_sample": false,
  "early_stopping": false,
  "num_beams": 1,
  "num_beam_groups": 1,
  "diversity_penalty": 0.0,
  "temperature": 1.0,
  "top_k": 50,
  "top_p": 1.0,
  "typical_p": 1.0,
  "repetition_penalty": 1.0,
  "length_penalty": 1.0,
  "no_repeat_ngram_size": 0,
  "encoder_no_repeat_ngram_size": 0,
  "bad_words_ids": null,
  "num_return_sequences": 1,
  "output_scores": false,
  "return_dict_in_generate": false,
  "forced_bos_token_id": null,
  "forced_eos_token_id": null,
  "remove_invalid_values": false,
  "exponential_decay_length_penalty": null,
  "suppress_tokens": null,
  "begin_suppress_tokens": null,
  "architectures": [
    "PaluLlamaForCausalLM"
  ],
  "finetuning_task": null,
  "id2label": {
    "0": "LABEL_0",
    "1": "LABEL_1"
  },
  "label2id": {
    "LABEL_0": 0,
    "LABEL_1": 1
  },
  "tokenizer_class": null,
  "prefix": null,
  "bos_token_id": 128000,
  "pad_token_id": null,
  "eos_token_id": 128009,
  "sep_token_id": null,
  "decoder_start_token_id": null,
  "task_specific_params": null,
  "problem_type": null,
  "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
  "transformers_version": "4.37.2",
  "model_type": "palullama",
  "head_wise_ranks": {
    "model.layers.0.self_attn.k_proj": [
      160,
      160
    ],
    "model.layers.0.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.1.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.1.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.2.self_attn.k_proj": [
      288,
      288
    ],
    "model.layers.2.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.3.self_attn.k_proj": [
      288,
      288
    ],
    "model.layers.3.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.4.self_attn.k_proj": [
      288,
      288
    ],
    "model.layers.4.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.5.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.5.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.6.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.6.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.7.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.7.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.8.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.8.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.9.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.9.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.10.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.10.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.11.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.11.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.12.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.12.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.13.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.13.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.14.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.14.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.15.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.15.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.16.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.16.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.17.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.17.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.18.self_attn.k_proj": [
      160,
      160
    ],
    "model.layers.18.self_attn.v_proj": [
      416,
      416
    ],
    "model.layers.19.self_attn.k_proj": [
      160,
      160
    ],
    "model.layers.19.self_attn.v_proj": [
      480,
      480
    ],
    "model.layers.20.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.20.self_attn.v_proj": [
      480,
      480
    ],
    "model.layers.21.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.21.self_attn.v_proj": [
      480,
      480
    ],
    "model.layers.22.self_attn.k_proj": [
      160,
      160
    ],
    "model.layers.22.self_attn.v_proj": [
      416,
      416
    ],
    "model.layers.23.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.23.self_attn.v_proj": [
      384,
      384
    ],
    "model.layers.24.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.24.self_attn.v_proj": [
      416,
      416
    ],
    "model.layers.25.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.25.self_attn.v_proj": [
      416,
      416
    ],
    "model.layers.26.self_attn.k_proj": [
      192,
      192
    ],
    "model.layers.26.self_attn.v_proj": [
      448,
      448
    ],
    "model.layers.27.self_attn.k_proj": [
      256,
      256
    ],
    "model.layers.27.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.28.self_attn.k_proj": [
      288,
      288
    ],
    "model.layers.28.self_attn.v_proj": [
      480,
      480
    ],
    "model.layers.29.self_attn.k_proj": [
      352,
      352
    ],
    "model.layers.29.self_attn.v_proj": [
      480,
      480
    ],
    "model.layers.30.self_attn.k_proj": [
      224,
      224
    ],
    "model.layers.30.self_attn.v_proj": [
      512,
      512
    ],
    "model.layers.31.self_attn.k_proj": [
      416,
      416
    ],
    "model.layers.31.self_attn.v_proj": [
      512,
      512
    ]
  },
  "original_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct"
}