| { |
| "config": null, |
| "condition": "transferred", |
| "target_model": "mistralai/Mistral-7B-v0.3", |
| "source_memory": "/scratch/project_462001050/mingyl/llm-memory-transfer/publish_bundle/models/hf_repos/olaresearchx-xmemtransfer-llama2-source-20m/memory.pt", |
| "memory_config": "/scratch/project_462001050/mingyl/llm-memory-transfer/publish_bundle/models/hf_repos/olaresearchx-xmemtransfer-llama2-source-20m/memory_config.json", |
| "init_adaptor": null, |
| "train_generated_branch_only": false, |
| "dual_reader_mode": "both", |
| "max_tokens": 20000000, |
| "seq_len": 2048, |
| "batch_size": 1, |
| "lr": 0.0003, |
| "warmup_steps": 200, |
| "seed": 42, |
| "output_dir": "/scratch/project_462001493/mingyl/ATHENA/run/tri_advantage_formal_20260908T133625Z/reader_20m_train", |
| "log_every": 50, |
| "eval_every": 2000, |
| "validation_max_tokens": 2000000, |
| "injection_layers": "2:10", |
| "adaptor_branches": 4, |
| "architecture": "generative", |
| "reader_type": "cross_attention", |
| "generator_cue_source": "hybrid", |
| "generator_num_latents": 4, |
| "generator_hidden_size": 256, |
| "generator_layers": 2, |
| "generator_heads": 4, |
| "generator_cue_window": 3, |
| "generator_adaptive_router": true, |
| "generator_router_hidden_size": 64, |
| "generator_router_semantic_size": 64, |
| "generator_source_adapter_rank": 16, |
| "generator_router_expert_mode": "residual", |
| "generator_fusion_type": "tri_reader", |
| "memory_dim": 512, |
| "canon_mode": "word_boundary", |
| "corpus": "wikipedia-2021", |
| "corpus_subset": "hq-dqa", |
| "wikipedia2021_dataset": null, |
| "wikipedia2021_source_tokenizer": "mistralai/Mistral-7B-v0.3", |
| "wikipedia2021_require_tokenizer_match": true, |
| "early_stopping_patience": 0, |
| "grad_accum_steps": 1, |
| "gradient_checkpointing": true, |
| "joint_engram_generated_router": false, |
| "joint_source_reader": false, |
| "joint_tri_reader": false, |
| "joint_tri_subset_reader": true, |
| "engram_source_loss_weight": 1.0, |
| "generated_source_loss_weight": 1.0, |
| "ge_source_loss_weight": 1.0, |
| "gh_source_loss_weight": 1.0, |
| "routed_source_loss_weight": 1.0, |
| "subset_routed_source_loss_weight": 1.0, |
| "tri_router_distillation_weight": 0.5, |
| "deterministic_engram_init_seed": 42, |
| "router_lr": 0.001, |
| "router_start_tokens": 2000000, |
| "router_init_alpha": 0.95, |
| "generated_residual_penalty": 0.0, |
| "skip_final_test_eval": true, |
| "adaptor_dir": "/scratch/project_462001493/mingyl/ATHENA/run/tri_memory_joint_softsubset_20260903T053716Z/tri_20m_train", |
| "adaptor_checkpoint": null, |
| "router_hidden_size": 64, |
| "router_semantic_size": 64, |
| "router_temperature": 1.0, |
| "span_lengths": [ |
| 1, |
| 4, |
| 8, |
| 16, |
| 32 |
| ], |
| "teacher_temperature": 0.15, |
| "advantage_weight_floor": 0.01, |
| "advantage_weight_cap": 2.0, |
| "router_thresholds": [ |
| 0.5, |
| 0.55, |
| 0.6, |
| 0.65, |
| 0.7, |
| 0.75, |
| 0.8, |
| 0.85, |
| 0.9, |
| 0.95 |
| ], |
| "advantage_thresholds": [ |
| 0.0, |
| 0.01, |
| 0.02, |
| 0.05, |
| 0.1, |
| 0.2, |
| 0.3, |
| 0.5 |
| ], |
| "candidate_space": "sources", |
| "advantage_threshold": 0.0, |
| "confidence_threshold": 0.5, |
| "advantage_max_scale": 1.0, |
| "advantage_regression_weight": 1.0, |
| "advantage_confidence_weight": 0.25, |
| "advantage_reader": { |
| "enabled": true, |
| "candidates": "sources", |
| "threshold": 0.0, |
| "confidence_threshold": 0.5, |
| "temperature": 0.15, |
| "max_scale": 1.0 |
| }, |
| "deployment_reader_mode": "tri_advantage_routed", |
| "router_strategy": "wikipedia_counterfactual_tri_advantage_distillation", |
| "router_training_data": "wikipedia-2021-causal-next-token-only", |
| "downstream_training_examples": 0, |
| "source_expert_checkpoint": "/scratch/project_462001493/mingyl/ATHENA/run/tri_memory_joint_softsubset_20260903T053716Z/tri_20m_train/adaptor_best.pt", |
| "experts_frozen": true, |
| "best_step": 4000 |
| } |
