config.json

{
  "architectures": [
    "Dots1ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "dtype": "float32",
  "eos_token_id": 151643,
  "first_k_dense_replace": 0,
  "hidden_act": "gelu",
  "hidden_size": 32,
  "initializer_range": 0.02,
  "intermediate_size": 32,
  "is_decoder": true,
  "layer_types": [
    "full_attention",
    "full_attention"
  ],
  "max_position_embeddings": 512,
  "max_window_layers": 62,
  "model_type": "dots1",
  "moe_intermediate_size": 16,
  "n_group": 1,
  "n_routed_experts": 8,
  "n_shared_experts": 1,
  "norm_topk_prob": false,
  "num_attention_heads": 2,
  "num_experts_per_tok": 8,
  "num_hidden_layers": 2,
  "num_key_value_heads": 2,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_parameters": {
    "rope_theta": 10000.0,
    "rope_type": "default"
  },
  "routed_scaling_factor": 1.0,
  "sliding_window": 4096,
  "tie_word_embeddings": false,
  "topk_group": 1,
  "transformers_version": "5.16.0.dev0",
  "use_cache": true,
  "vocab_size": 151659
}