config.json

{
  "architectures": [
    "OlmoForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attn_implementation": "flash_attention_2",
  "clip_qkv": 8.0,
  "dtype": "float32",
  "eos_token_id": 50279,
  "hidden_act": "silu",
  "hidden_size": 512,
  "initializer_range": 0.02,
  "intermediate_size": 2048,
  "max_position_embeddings": 2048,
  "model_type": "olmo",
  "num_attention_heads": 8,
  "num_hidden_layers": 8,
  "num_key_value_heads": 8,
  "pad_token_id": 1,
  "rope_parameters": {
    "rope_theta": 10000.0,
    "rope_type": "default"
  },
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": true,
  "transformers_version": "4.57.3",
  "use_cache": false,
  "vocab_size": 50280
}