{
  "step": 244,
  "val_bpb": 0.612848621740378,
  "model_config": {
    "sequence_len": 4096,
    "vocab_size": 32768,
    "n_layer": 12,
    "n_head": 4,
    "n_kv_head": 4,
    "n_embd": 768,
    "window_pattern": "SSSL"
  },
  "user_config": {
    "run": "dummy",
    "device_type": "",
    "model_tag": "simple_20261104_nodpo_smoltalk_multiturn_codexrl",
    "model_step": null,
    "load_source": "sft",
    "output_tag": "simple_20261104_nodpo_smoltalk_multiturn_codexrl_multiturn",
    "load_optimizer": 0,
    "num_iterations": -1,
    "max_seq_len": 4096,
    "device_batch_size": 4,
    "total_batch_size": 1048576,
    "embedding_lr": null,
    "unembedding_lr": null,
    "matrix_lr": null,
    "init_lr_frac": 0.8,
    "warmup_ratio": 0.0,
    "warmdown_ratio": 0.5,
    "final_lr_frac": 0.0,
    "eval_every": 200,
    "eval_tokens": 131072,
    "chatcore_every": 200,
    "chatcore_max_cat": -1,
    "chatcore_max_sample": 24,
    "mmlu_epochs": 3,
    "gsm8k_epochs": 4,
    "chat_only": false,
    "multiturn_only": true,
    "multiturn_subset": "chats_v2_k200000",
    "sft_warmup": false,
    "tool_mix": false,
    "smoltalk_partition": "all"
  }
}