config.json

{
  "_attn_implementation_autoset": true,
  "_name_or_path": "katuni4ka/tiny-random-falcon-40b",
  "activation": "gelu",
  "alibi": false,
  "apply_residual_connection_post_layernorm": false,
  "architectures": [
    "FalconForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bias": false,
  "bos_token_id": null,
  "eos_token_id": 11,
  "ffn_hidden_size": 512,
  "hidden_dropout": 0.0,
  "hidden_size": 128,
  "initializer_range": 0.02,
  "layer_norm_epsilon": 1e-05,
  "max_position_embeddings": 2048,
  "model_type": "falcon",
  "multi_query": true,
  "new_decoder_architecture": true,
  "num_attention_heads": 16,
  "num_hidden_layers": 2,
  "num_kv_heads": 8,
  "num_ln_in_parallel_attn": 2,
  "parallel_attn": true,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "torch_dtype": "float32",
  "transformers_version": "4.46.0",
  "use_cache": false,
  "vocab_size": 65024
}