appvoid commited on
Commit
56533fc
·
verified ·
1 Parent(s): e1ec589

Update TensorBoard training metrics

Browse files
metrics/void-byte-v1/1791039222312944374-63c10d34/eval.jsonl ADDED
@@ -0,0 +1 @@
 
 
1
+ {"step": 605000, "seed_holdout": {"answer_byte_loss": 0.09305128062313253, "teacher_forced_byte_accuracy": 0.9696969696969697, "generation_exact_match": 0.5833333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.073719745212131, "teacher_forced_byte_accuracy": 0.3, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}}
metrics/void-byte-v1/eval.jsonl CHANGED
@@ -1,9 +1 @@
1
- {"step": 560000, "seed_holdout": {"answer_byte_loss": 0.8940724665590096, "teacher_forced_byte_accuracy": 0.6875, "generation_exact_match": 0.25, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.6826861487494575, "teacher_forced_byte_accuracy": 0.2, "generation_exact_match": 0.0, "generation_truncated": 2, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_xor_constraints", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_algebra_invariant", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_conservation_transfer", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_causal_delay", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_simulated_multimodal_motion", "surface": "symbolic", "exact": true, "generated_bytes": 3, "target_bytes": 3}, {"group": "seed_holdout", "family": "gen_counterfactual_intervention", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_abduction_set", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_corrupted_rule_robustness", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_cross_faculty_transfer", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_identity_copy", "surface": "symbolic", "exact": false, "generated_bytes": 4, "target_bytes": 7}, {"group": "seed_holdout", "family": "gen_equality_detection", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_odd_one_out", "surface": "symbolic", "exact": false, "generated_bytes": 3, "target_bytes": 2}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
2
- {"step": 565000, "seed_holdout": {"answer_byte_loss": 0.05673352440657175, "teacher_forced_byte_accuracy": 0.9891304347826086, "generation_exact_match": 0.8333333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.3207723405626086, "teacher_forced_byte_accuracy": 0.3111111111111111, "generation_exact_match": 0.0, "generation_truncated": 2, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_majority_symbol", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_minority_symbol", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_parity_count", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_set_membership_basic", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_adjacency_1d", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_run_length_count", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_deduplicate_sequence", "surface": "symbolic", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_first_last_relation", "surface": "symbolic", "exact": false, "generated_bytes": 3, "target_bytes": 3}, {"group": "seed_holdout", "family": "gen_translate_right", "surface": "symbolic", "exact": true, "generated_bytes": 40, "target_bytes": 40}, {"group": "seed_holdout", "family": "gen_translate_down", "surface": "symbolic", "exact": true, "generated_bytes": 47, "target_bytes": 47}, {"group": "seed_holdout", "family": "gen_reflect_h", "surface": "symbolic", "exact": true, "generated_bytes": 47, "target_bytes": 47}, {"group": "seed_holdout", "family": "gen_reflect_v", "surface": "symbolic", "exact": true, "generated_bytes": 29, "target_bytes": 29}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 20, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
3
- {"step": 570000, "seed_holdout": {"answer_byte_loss": 0.06971308760248221, "teacher_forced_byte_accuracy": 0.987012987012987, "generation_exact_match": 0.6666666666666666, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.5129309548272025, "teacher_forced_byte_accuracy": 0.28888888888888886, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_rotate90", "surface": "symbolic", "exact": true, "generated_bytes": 41, "target_bytes": 41}, {"group": "seed_holdout", "family": "gen_rotate180", "surface": "symbolic", "exact": false, "generated_bytes": 41, "target_bytes": 40}, {"group": "seed_holdout", "family": "gen_transpose", "surface": "symbolic", "exact": true, "generated_bytes": 25, "target_bytes": 25}, {"group": "seed_holdout", "family": "gen_crop_bbox", "surface": "symbolic", "exact": false, "generated_bytes": 7, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_bbox_outline", "surface": "symbolic", "exact": true, "generated_bytes": 61, "target_bytes": 61}, {"group": "seed_holdout", "family": "gen_recolor_largest", "surface": "symbolic", "exact": true, "generated_bytes": 61, "target_bytes": 61}, {"group": "seed_holdout", "family": "gen_recolor_smallest", "surface": "symbolic", "exact": true, "generated_bytes": 25, "target_bytes": 25}, {"group": "seed_holdout", "family": "gen_extract_largest", "surface": "symbolic", "exact": true, "generated_bytes": 11, "target_bytes": 11}, {"group": "seed_holdout", "family": "gen_complete_symmetry", "surface": "symbolic", "exact": true, "generated_bytes": 47, "target_bytes": 47}, {"group": "seed_holdout", "family": "gen_tile_motif", "surface": "symbolic", "exact": true, "generated_bytes": 25, "target_bytes": 25}, {"group": "seed_holdout", "family": "gen_count_to_bar", "surface": "symbolic", "exact": false, "generated_bytes": 9, "target_bytes": 11}, {"group": "seed_holdout", "family": "gen_remove_noise", "surface": "symbolic", "exact": false, "generated_bytes": 30, "target_bytes": 30}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 22, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 7, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
4
- {"step": 575000, "seed_holdout": {"answer_byte_loss": 0.06719702047352888, "teacher_forced_byte_accuracy": 0.9795918367346939, "generation_exact_match": 0.8333333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.1740833282470704, "teacher_forced_byte_accuracy": 0.37777777777777777, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_fill_holes", "surface": "symbolic", "exact": true, "generated_bytes": 40, "target_bytes": 40}, {"group": "seed_holdout", "family": "gen_conditional_transform", "surface": "symbolic", "exact": false, "generated_bytes": 46, "target_bytes": 47}, {"group": "seed_holdout", "family": "gen_two_rule_comp", "surface": "symbolic", "exact": true, "generated_bytes": 25, "target_bytes": 25}, {"group": "seed_holdout", "family": "gen_three_rule_comp", "surface": "symbolic", "exact": false, "generated_bytes": 7, "target_bytes": 9}, {"group": "seed_holdout", "family": "gen_sort_rows", "surface": "symbolic", "exact": true, "generated_bytes": 35, "target_bytes": 35}, {"group": "seed_holdout", "family": "gen_negative_space", "surface": "symbolic", "exact": true, "generated_bytes": 41, "target_bytes": 41}, {"group": "seed_holdout", "family": "gen_occlusion_complete", "surface": "symbolic", "exact": true, "generated_bytes": 35, "target_bytes": 35}, {"group": "seed_holdout", "family": "gen_single_fact", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_two_hop", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_multi_hop", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_inverse", "surface": "instruction_json", "exact": true, "generated_bytes": 5, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_transitivity", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 38, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
5
- {"step": 580000, "seed_holdout": {"answer_byte_loss": 0.04150199038849678, "teacher_forced_byte_accuracy": 1.0, "generation_exact_match": 1.0, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.4243011474609375, "teacher_forced_byte_accuracy": 0.24444444444444444, "generation_exact_match": 0.0, "generation_truncated": 2, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_symmetric", "surface": "instruction_json", "exact": true, "generated_bytes": 5, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_temporal_order", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_relative_rank", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_relation_comp", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_exactly_one", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_at_least", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_at_most", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_quantifier", "surface": "instruction_json", "exact": true, "generated_bytes": 5, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_ordering_csp", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_assignment_matching", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_small_sat", "surface": "instruction_json", "exact": true, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_contradiction", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
6
- {"step": 585000, "seed_holdout": {"answer_byte_loss": 0.1353585021445957, "teacher_forced_byte_accuracy": 0.9651162790697675, "generation_exact_match": 0.8333333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.1825502607557508, "teacher_forced_byte_accuracy": 0.37777777777777777, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_kv_lookup", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_role_filler", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_assoc_distractors", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_nested_lookup", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_mutable_dictionary", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_queue", "surface": "instruction_json", "exact": true, "generated_bytes": 18, "target_bytes": 18}, {"group": "seed_holdout", "family": "gen_deque", "surface": "instruction_json", "exact": true, "generated_bytes": 14, "target_bytes": 14}, {"group": "seed_holdout", "family": "gen_delayed_recall", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_register_swap", "surface": "instruction_json", "exact": false, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_register_update", "surface": "instruction_json", "exact": true, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_conditional_branch", "surface": "instruction_json", "exact": true, "generated_bytes": 12, "target_bytes": 12}, {"group": "seed_holdout", "family": "gen_loop_counter", "surface": "instruction_json", "exact": false, "generated_bytes": 10, "target_bytes": 14}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 5, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
7
- {"step": 590000, "seed_holdout": {"answer_byte_loss": 0.5058835760422357, "teacher_forced_byte_accuracy": 0.8493150684931506, "generation_exact_match": 0.5, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.2474500444200305, "teacher_forced_byte_accuracy": 0.3333333333333333, "generation_exact_match": 0.0, "generation_truncated": 2, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_multi_register", "surface": "instruction_json", "exact": false, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_program_trace", "surface": "instruction_json", "exact": false, "generated_bytes": 42, "target_bytes": 42}, {"group": "seed_holdout", "family": "gen_stack_expr", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_prefix_expr", "surface": "instruction_json", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_postfix_expr", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_bracket_balance", "surface": "instruction_json", "exact": true, "generated_bytes": 5, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_bracket_depth", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_multiply", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_divide", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_remainder", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_modular", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_magnitude", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 37, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
8
- {"step": 595000, "seed_holdout": {"answer_byte_loss": 0.1765477716298147, "teacher_forced_byte_accuracy": 0.9519230769230769, "generation_exact_match": 0.5833333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.2510670132107204, "teacher_forced_byte_accuracy": 0.2777777777777778, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_ratio", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_fraction_equiv", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_difference", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_linear", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_affine", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_unit_conversion", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_relative_coords", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_line_of_sight", "surface": "instruction_json", "exact": false, "generated_bytes": 6, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_containment", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_collision", "surface": "instruction_json", "exact": true, "generated_bytes": 35, "target_bytes": 35}, {"group": "seed_holdout", "family": "gen_toroidal", "surface": "instruction_json", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_falling", "surface": "instruction_json", "exact": true, "generated_bytes": 34, "target_bytes": 34}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 30, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 19, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
9
- {"step": 600000, "seed_holdout": {"answer_byte_loss": 0.09662931144106178, "teacher_forced_byte_accuracy": 0.9615384615384616, "generation_exact_match": 0.9166666666666666, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.2137988620334204, "teacher_forced_byte_accuracy": 0.4111111111111111, "generation_exact_match": 0.0, "generation_truncated": 2, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_obstacle_motion", "surface": "instruction_json", "exact": true, "generated_bytes": 14, "target_bytes": 14}, {"group": "seed_holdout", "family": "gen_path_keys", "surface": "instruction_json", "exact": false, "generated_bytes": 12, "target_bytes": 12}, {"group": "seed_holdout", "family": "gen_dynamic_plan", "surface": "instruction_json", "exact": true, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_direct_cause", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_confounding", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_competing_hypotheses", "surface": "instruction_json", "exact": true, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_noisy_evidence", "surface": "instruction_json", "exact": true, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_active_disambiguation", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_qual_bayes", "surface": "instruction_json", "exact": true, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_causal_intervention", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_causal_abduction", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_rule_change", "surface": "instruction_json", "exact": true, "generated_bytes": 10, "target_bytes": 10}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
 
1
+ {"step": 605000, "seed_holdout": {"answer_byte_loss": 0.09305128062313253, "teacher_forced_byte_accuracy": 0.9696969696969697, "generation_exact_match": 0.5833333333333334, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 2.073719745212131, "teacher_forced_byte_accuracy": 0.3, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_anomaly_cause", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_selection_sort", "surface": "instruction_json", "exact": false, "generated_bytes": 50, "target_bytes": 50}, {"group": "seed_holdout", "family": "gen_insertion_sort", "surface": "instruction_json", "exact": true, "generated_bytes": 18, "target_bytes": 18}, {"group": "seed_holdout", "family": "gen_minmax_scan", "surface": "instruction_json", "exact": true, "generated_bytes": 20, "target_bytes": 20}, {"group": "seed_holdout", "family": "gen_linear_search", "surface": "instruction_json", "exact": true, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_binary_search", "surface": "instruction_json", "exact": false, "generated_bytes": 8, "target_bytes": 10}, {"group": "seed_holdout", "family": "gen_string_match", "surface": "instruction_json", "exact": false, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_bfs_order", "surface": "instruction_json", "exact": true, "generated_bytes": 14, "target_bytes": 14}, {"group": "seed_holdout", "family": "gen_dfs_order", "surface": "instruction_json", "exact": true, "generated_bytes": 12, "target_bytes": 12}, {"group": "seed_holdout", "family": "gen_topo_sort", "surface": "instruction_json", "exact": true, "generated_bytes": 14, "target_bytes": 14}, {"group": "seed_holdout", "family": "gen_union_find", "surface": "instruction_json", "exact": false, "generated_bytes": 6, "target_bytes": 5}, {"group": "seed_holdout", "family": "gen_shortest_path", "surface": "instruction_json", "exact": false, "generated_bytes": 10, "target_bytes": 8}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
 
 
 
 
 
 
 
 
runs/void-byte-v1/events.out.tfevents.1791039222.f2a393d4-e1e4-407b-a6aa-034aa2ada5ac-vwfgg.353.0.1791039222312944374-63c10d34 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e9bdb81cfb097c4ddf126f6c6bf56636e1f9cbb27e9f2b757532d390fe25050b
3
- size 12873
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:241415ea1c589a80401ab37cf6489785c002a082da7aee7f181c97e6f4c2c2be
3
+ size 13656