{
  "parameters": 44996036,
  "neural_core_parameters": 4996035,
  "ngram_parameters": 40000001,
  "architecture": {
    "vocab_size": 2048,
    "hidden_size": 256,
    "intermediate_size": 1488,
    "num_hidden_layers": 3,
    "num_attention_heads": 8,
    "num_key_value_heads": 1,
    "head_dim": 32,
    "ngram_buckets": 156250,
    "loop_schedule": [
      0,
      1,
      1,
      2
    ],
    "refresh_kernel_size": 9,
    "max_position_embeddings": 4096,
    "rope_theta": 100000.0,
    "rms_norm_eps": 1e-05,
    "initializer_range": 0.02,
    "use_cache": false,
    "transformers_version": "5.16.1",
    "architectures": null,
    "output_hidden_states": false,
    "return_dict": true,
    "dtype": null,
    "chunk_size_feed_forward": 0,
    "is_encoder_decoder": false,
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1"
    },
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1
    },
    "problem_type": null,
    "_name_or_path": "",
    "tie_word_embeddings": true,
    "bos_token_id": 1,
    "eos_token_id": 2,
    "pad_token_id": 0,
    "unk_token_id": 3,
    "model_type": "cherry_alpha",
    "output_attentions": false
  },
  "training_percent": 100,
  "step": 38147,
  "total_steps": 38147,
  "tokens_seen": 20000014336,
  "target_tokens": 20000000000,
  "trained_token_target": 20000014336,
  "tokens_per_step": 524288,
  "local_batch": 32,
  "effective_global_batch": 128,
  "world_size": 4,
  "gpu_name": "NVIDIA H200",
  "optimizer": "Muon + AdamW",
  "muon_peak_lr": 0.02,
  "adamw_peak_lr": 0.002,
  "ngram_peak_lr": 0.001,
  "architecture_revision": "2d06c38b56820561813f800da2f896ad64824877",
  "tokenizer_revision": "2d06c38b56820561813f800da2f896ad64824877",
  "dataset_id": "HuggingFaceFW/fineweb-edu",
  "dataset_config": "sample-100BT",
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
  "elapsed_seconds": 8416.391685962677
}