Update TensorBoard training metrics

metrics/void-byte-v1/1791300814123050567-72be911d/eval.jsonl

@@ -1 +1,2 @@
 {"step": 805000, "seed_holdout": {"answer_byte_loss": 0.0020957421402272303, "teacher_forced_byte_accuracy": 1.0, "generation_exact_match": 1.0, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 1.5949350992838542, "teacher_forced_byte_accuracy": 0.3888888888888889, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}}
+{"step": 810000, "seed_holdout": {"answer_byte_loss": 0.5795302560527884, "teacher_forced_byte_accuracy": 0.8369565217391305, "generation_exact_match": 0.4166666666666667, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 1.6899066077338325, "teacher_forced_byte_accuracy": 0.37777777777777777, "generation_exact_match": 0.0, "generation_truncated": 0, "episodes": 3}}

metrics/void-byte-v1/eval.jsonl

@@ -1 +1,2 @@
 {"step": 805000, "seed_holdout": {"answer_byte_loss": 0.0020957421402272303, "teacher_forced_byte_accuracy": 1.0, "generation_exact_match": 1.0, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 1.5949350992838542, "teacher_forced_byte_accuracy": 0.3888888888888889, "generation_exact_match": 0.0, "generation_truncated": 1, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_reference_resolution", "surface": "instruction_json", "exact": true, "generated_bytes": 10, "target_bytes": 10}, {"group": "seed_holdout", "family": "gen_instruction_override", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_output_schema", "surface": "instruction_json", "exact": true, "generated_bytes": 10, "target_bytes": 10}, {"group": "seed_holdout", "family": "gen_boolean_formula", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_set_expression", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_bounded_arithmetic", "surface": "instruction_json", "exact": true, "generated_bytes": 3, "target_bytes": 3}, {"group": "seed_holdout", "family": "gen_text_span_extract", "surface": "instruction_json", "exact": true, "generated_bytes": 4, "target_bytes": 4}, {"group": "seed_holdout", "family": "gen_counterfactual_state", "surface": "instruction_json", "exact": true, "generated_bytes": 26, "target_bytes": 26}, {"group": "seed_holdout", "family": "gen_ambiguity_detection", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_multi_step_word_problem", "surface": "instruction_json", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_compare_entities", "surface": "instruction_json", "exact": true, "generated_bytes": 8, "target_bytes": 8}, {"group": "seed_holdout", "family": "gen_missing_information", "surface": "instruction_json", "exact": true, "generated_bytes": 10, "target_bytes": 10}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 18, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 12, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 192, "target_bytes": 16}]}
+{"step": 810000, "seed_holdout": {"answer_byte_loss": 0.5795302560527884, "teacher_forced_byte_accuracy": 0.8369565217391305, "generation_exact_match": 0.4166666666666667, "generation_truncated": 0, "episodes": 12}, "composition_ood": {"answer_byte_loss": 1.6899066077338325, "teacher_forced_byte_accuracy": 0.37777777777777777, "generation_exact_match": 0.0, "generation_truncated": 0, "episodes": 3}, "families": [{"group": "seed_holdout", "family": "gen_string_normalization", "surface": "instruction_json", "exact": true, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_list_zip", "surface": "instruction_json", "exact": true, "generated_bytes": 3, "target_bytes": 3}, {"group": "seed_holdout", "family": "gen_entailment_closed_world", "surface": "instruction_json", "exact": true, "generated_bytes": 7, "target_bytes": 7}, {"group": "seed_holdout", "family": "gen_periodic", "surface": "symbolic", "exact": false, "generated_bytes": 1, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_mirror", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 9}, {"group": "seed_holdout", "family": "gen_cardinality_grounding", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_successor_chain", "surface": "symbolic", "exact": false, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_motif_growth", "surface": "symbolic", "exact": true, "generated_bytes": 37, "target_bytes": 37}, {"group": "seed_holdout", "family": "gen_latent_arithmetic", "surface": "symbolic", "exact": true, "generated_bytes": 2, "target_bytes": 2}, {"group": "seed_holdout", "family": "gen_sequence_transform", "surface": "symbolic", "exact": false, "generated_bytes": 6, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_binary_set_logic", "surface": "symbolic", "exact": false, "generated_bytes": 4, "target_bytes": 6}, {"group": "seed_holdout", "family": "gen_substitution_cipher", "surface": "symbolic", "exact": false, "generated_bytes": 13, "target_bytes": 10}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 12, "target_bytes": 18}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 18, "target_bytes": 56}, {"group": "composition_ood", "family": "compositional_execution", "surface": "symbolic", "exact": false, "generated_bytes": 13, "target_bytes": 16}]}

runs/void-byte-v1/events.out.tfevents.1791300814.0c9d201b-2dbc-40fb-b739-4ee2350a0f54-7b2tx.840.0.1791300814123050567-72be911d

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1dd5f35197d52e96eedc196895568978d24e029395fd338c5bc99f359bca08bb
-size 212748
+oid sha256:379f086bab2499dc7befb2b5f26bfa21abd1c8db0c750eeed806438851a3f139
+size 213531