config.json

{
  "_name_or_path": "fxmarty/tiny-llama-fast-tokenizer",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 0,
  "eos_token_id": 1,
  "head_dim": 4,
  "hidden_act": "silu",
  "hidden_size": 16,
  "initializer_range": 0.02,
  "intermediate_size": 64,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 4,
  "num_hidden_layers": 2,
  "num_key_value_heads": 4,
  "pad_token_id": -1,
  "pretraining_tp": 1,
  "quantization_config": {
    "algo_config": null,
    "exclude": [
      "lm_head"
    ],
    "export": {
      "kv_cache_group": [],
      "pack_method": "reorder",
      "weight_format": "real_quantized",
      "weight_merge_groups": null
    },
    "global_quant_config": {
      "bias": null,
      "input_tensors": {
        "ch_axis": null,
        "dtype": "fp8_e4m3",
        "group_size": null,
        "is_dynamic": false,
        "observer_cls": "PerTensorMinMaxObserver",
        "qscheme": "per_tensor",
        "round_method": null,
        "scale_type": null,
        "symmetric": null
      },
      "output_tensors": null,
      "target_device": null,
      "weight": {
        "ch_axis": null,
        "dtype": "fp8_e4m3",
        "group_size": null,
        "is_dynamic": false,
        "observer_cls": "PerTensorMinMaxObserver",
        "qscheme": "per_tensor",
        "round_method": null,
        "scale_type": null,
        "symmetric": null
      }
    },
    "layer_quant_config": {},
    "layer_type_quant_config": {},
    "pack_method": "reorder",
    "quant_method": "quark",
    "quant_mode": "eager_mode"
  },
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.45.2",
  "use_cache": true,
  "vocab_size": 32000
}