| { |
| "num_rollouts": 4, |
| "K": 4, |
| "max_iterations": 2, |
| "memory_span_abs": 1792, |
| "memory_span_traj": 1792, |
| "temperature": 1.0, |
| "ar_search": false, |
| "response_only_abs": false, |
| "alpha_info_gain": 10.0, |
| "alpha_abs": 0.1, |
| "alpha_soft_zipf": 1.0, |
| "alpha_ortho": 0.0, |
| "alpha_anchor": 0.0, |
| "alpha_jacobi": 0.0, |
| "decay": 0.8, |
| "target_vocab_util": 0.8, |
| "min_abs_ppl": 0.0, |
| "zipf_alpha": 1.0, |
| "lr": 2e-05, |
| "emb_lr_mult": 1.0, |
| "weight_decay": 0.01, |
| "warmup_steps": 100, |
| "cooldown_frac": 0.4, |
| "max_grad_norm": 1.0, |
| "vq_abs_pretrain_steps": 0, |
| "vq_abs_pretrain_lr": 0.001, |
| "vq_abs_pretrain_layer": -1, |
| "vq_abs_pretrain_batch_size": 256, |
| "vq_abs_pretrain_target_vectors": 20000, |
| "batch_size": 64, |
| "gradient_accumulation_steps": 1, |
| "num_epochs": 20, |
| "emb_warmup_steps": 0, |
| "log_every": 50, |
| "eval_every": 156, |
| "save_every": 999999, |
| "eval_samples": 100, |
| "output_dir": "ckpt/sweep/as_baseline_10K_1L2H256d", |
| "eval_K": 4, |
| "alpha_traj": 0.0, |
| "corrupt_method": "shuffle", |
| "corrupt_ratio": 0.3, |
| "alpha_contrastive": 1.0, |
| "gamma_contrastive": 0.5, |
| "alpha_masked_traj": 0.0, |
| "mask_nl_ratio": 0.3, |
| "mask_nl_mode": "fixed", |
| "mask_nl_fixed_id": 0, |
| "use_ste": true, |
| "n_inner": 1, |
| "random_K": null, |
| "strip_suffix": null, |
| "compress_prefix": null, |
| "random_mem_span": null, |
| "warmup_ratio": 0.03, |
| "beta2": 0.999, |
| "seed": 42, |
| "n_digits": 6, |
| "n_layer": 1, |
| "n_head": 2, |
| "n_embd": 256, |
| "ops": "add_sub", |
| "abs_vocab": 0, |
| "dataset_size": 10000, |
| "mode": "baseline", |
| "device": "cuda", |
| "push_to_hub": true, |
| "no_wandb": false, |
| "n_params": 78691840, |
| "run_name": "add_sub_baseline_10K_1L2H256d", |
| "git_commit": "1d5a160e16a5070d61b881494e832aa88149b15c", |
| "timestamp": "2026-04-15T05:21:44.700980+00:00", |
| "tokenizer": "Qwen/Qwen3-0.6B", |
| "dataset_repo": "thoughtworks/arithmetic-sorl-data", |
| "dataset_config": "add_sub_6digit", |
| "train_dataset": "fixed_train/train_10K_seed42.pt", |
| "model_repo": "thoughtworks/arithmetic-sorl", |
| "trainer_version": "sft", |
| "wandb_run_id": "s9ep067n", |
| "wandb_url": "https://wandb.ai/nlp_and_interpretability/sorl-arithmetic/runs/s9ep067n", |
| "eval_final_dataset": "eval_sets/eval_add_sub_6d_N100_seed42.json", |
| "eval_epoch_dataset": "eval_sets/eval_add_sub_6d_N25_seed42.json", |
| "eval_hf_repo": "thoughtworks/arithmetic-sorl-data", |
| "final_accuracy": 0.0019230769230769232, |
| "sft_accuracy": 0.0019230769230769232, |
| "eval_method": "ArithmeticEvaluator", |
| "config_hash": "2e75ee47683a" |
| } |
