config.json

{
  "architectures": [
    "GPTOSSForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 50256,
  "eos_token_id": 50256,
  "hidden_size": 512,
  "intermediate_size": 1536,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 8192,
  "model_type": "gptossmini",
  "moe_top_k": 2,
  "num_attention_heads": 8,
  "num_experts": 8,
  "num_hidden_layers": 6,
  "num_key_value_heads": 2,
  "pad_token_id": 50256,
  "resid_dropout": 0.0,
  "rope_theta": 10000.0,
  "torch_dtype": "float32",
  "transformers_version": "4.55.2",
  "use_cache": true,
  "vocab_size": 48000
}