saves-bloom-cosine/config.json

{
  "apply_residual_connection_post_layernorm": false,
  "architectures": [
    "BloomForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "gelu",
  "hidden_dropout": 0.0,
  "hidden_size": 256,
  "initializer_range": 0.02,
  "intermediate_size": 1024,
  "layer_norm_epsilon": 1e-05,
  "model_type": "bloom",
  "n_head": 8,
  "n_layer": 2,
  "num_key_value_heads": 8,
  "pretraining_tp": 1,
  "slow_but_exact": false,
  "torch_dtype": "float32",
  "transformers_version": "4.42.4",
  "use_cache": true,
  "vocab_size": 2000
}