{
  "_name_or_path": "",
  "architectures": [
    "Olmo3ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": null,
  "eos_token_id": 100257,
  "hidden_act": "silu",
  "hidden_dropout": 0.0,
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "layer_types": [
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention"
  ],
  "max_position_embeddings": 8192,
  "max_sequence_length": 8192,
  "mlp_bias": false,
  "model_type": "olmo3",
  "num_attention_heads": 16,
  "num_hidden_layers": 16,
  "num_key_value_heads": 16,
  "olmo3_weight_decay": true,
  "pad_token_id": 100277,
  "padded_vocab_size": 100352,
  "qk_norm": true,
  "qk_norm_mode": "full_projection",
  "rms_norm_eps": 1e-06,
  "rope_full_precision": true,
  "rope_scaling": null,
  "rope_theta": 500000.0,
  "scale_embeds": false,
  "scale_output_layer_init": false,
  "sliding_window": 4096,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.57.6",
  "true_vocab_size": 100278,
  "truncated_normal_factor": 3.0,
  "use_cache": true,
  "vocab_size": 100278,
  "vocab_z_loss_coeff": 1e-05
}