question stringclasses 3
values | metadata stringclasses 3
values | task_source stringclasses 1
value | formatted_prompt listlengths 2 2 | responses_by_sample listlengths 1 1 | tokens_by_sample listlengths 1 1 | num_rounds_by_sample listlengths 1 1 | continuation_prompts_by_sample listlengths 1 1 | continuation_text_by_sample listlengths 1 1 | strategy_type stringclasses 1
value | termination_reason_by_sample listlengths 1 1 | correct_round_by_sample listlengths 1 1 | smart_termination_enabled bool 1
class | retry_count_by_sample listlengths 1 1 | answers_by_sample listlengths 1 1 | evals_by_sample listlengths 1 1 | eval_metadata_by_sample listlengths 1 1 | annotations_by_sample listlengths 1 1 | arithmetic_attempts_by_sample listlengths 1 1 | expression_attempts_by_sample listlengths 1 1 | subgoals_by_sample listlengths 1 1 | failure_markers_by_sample listlengths 1 1 | subgoal_windows_by_sample listlengths 1 1 | branches_by_sample listlengths 1 1 | extraction_stats_by_sample listlengths 1 1 | llm_supplement_stats_by_sample listlengths 1 1 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
"Using ALL of the numbers 12, 4, 7, 1, 2, 5, 21, 88, create an expression that equals 109. You must (...TRUNCATED) | "{\"numbers\": [12, 4, 7, 1, 2, 5, 21, 88], \"target\": 109, \"operators_used\": [\"/\", \"+\", \"/\(...TRUNCATED) | countdown | [{"content":"You are a helpful assistant that solves math problems by reasoning step-by-step. Show y(...TRUNCATED) | [["Okay, let's see. I need to use all the numbers 12, 4, 7, 1, 2, 5, 21, 88 exactly once with +, -, (...TRUNCATED) | [
32941
] | [
4
] | [["Continue exploring alternative solution strategies.","Continue exploring alternative solution str(...TRUNCATED) | [["Okay, let's see. I need to use all the numbers 12, 4, 7, 1, 2, 5, 21, 88 exactly once with +, -, (...TRUNCATED) | iterative | [
"budget_reached"
] | [
null
] | false | [
0
] | [[["88 + 21 + (12 - 7 - 5) * (4 - 2 - 1)"],["88 + 21 + (12 - 4 - 7 - 1) * (5 - 2)"],["88 + 21 + ((7 (...TRUNCATED) | [
[
[
true
],
[
true
],
[
true
],
[
true,
true
]
]
] | [[[{"answer_block":"88 + 21 + (12 - 7 - 5) * (4 - 2 - 1)","error":null,"final_answer":109.0,"is_corr(...TRUNCATED) | [
[]
] | [[{"all_operands":[109.0,88.0],"all_operators":["-"],"branch_id":0,"canonical_form":"109 - 88 = 21",(...TRUNCATED) | [[{"branch_id":0,"canonical_key":"expr:109-88=21","end_char":487,"end_char_in_round":487,"eval_value(...TRUNCATED) | [[{"canonical_key":"subgoal:21","confidence":0.85,"end_char":1268,"end_char_in_round":1268,"goal_val(...TRUNCATED) | [[{"end_char":4984,"end_char_in_round":4984,"marker_type":"not_helpful","round_idx":0,"span_text":"N(...TRUNCATED) | [[{"branch_id":0,"end_char":6754,"end_char_in_round":6754,"evidence_kind":"need","evidence_text":"ne(...TRUNCATED) | [[{"branch_id":0,"end_char":7501,"end_char_in_round":7501,"round_idx":0,"start_char":0,"start_char_i(...TRUNCATED) | [{"attempts_by_match_type":{"infix":526,"latex_frac":3,"multi_op_chain":113,"nested_paren":17,"paren(...TRUNCATED) | [{"candidates_proposed":483,"candidates_validated":715,"new_span_finds":85,"novel_key_finds":42,"spa(...TRUNCATED) |
"Using ALL of the numbers 25, 69, 58, 14, 55, 24, 36, 60, create an expression that equals 176. You (...TRUNCATED) | "{\"numbers\": [25, 69, 58, 14, 55, 24, 36, 60], \"target\": 176, \"operators_used\": [\"-\", \"+\",(...TRUNCATED) | countdown | [{"content":"You are a helpful assistant that solves math problems by reasoning step-by-step. Show y(...TRUNCATED) | [["Okay, let's see. I need to use all the numbers 25, 69, 58, 14, 55, 24, 36, 60 exactly once with +(...TRUNCATED) | [
39403
] | [
1
] | [
[]
] | [["Okay, let's see. I need to use all the numbers 25, 69, 58, 14, 55, 24, 36, 60 exactly once with +(...TRUNCATED) | iterative | [
"budget_reached"
] | [
null
] | false | [
0
] | [
[
[
"(69 - 55) * (25 - 14) + (60 - 58) * (36 / 24)"
]
]
] | [
[
[
false
]
]
] | [[[{"answer_block":"(69 - 55) * (25 - 14) + (60 - 58) * (36 / 24)","error":"Expression does not equa(...TRUNCATED) | [
[]
] | [[{"all_operands":[25.0,69.0],"all_operators":["+"],"branch_id":0,"canonical_form":"25 + 69 = 94","c(...TRUNCATED) | [[{"branch_id":0,"canonical_key":"expr:25+69=94","end_char":582,"end_char_in_round":582,"eval_value"(...TRUNCATED) | [[{"canonical_key":"subgoal:176","confidence":0.9,"end_char":320,"end_char_in_round":320,"goal_value(...TRUNCATED) | [[{"end_char":343,"end_char_in_round":343,"marker_type":"too_big","round_idx":0,"span_text":"too lar(...TRUNCATED) | [[{"branch_id":0,"end_char":711,"end_char_in_round":711,"evidence_kind":"goal","evidence_text":"get (...TRUNCATED) | [[{"branch_id":0,"end_char":50669,"end_char_in_round":50669,"round_idx":0,"start_char":0,"start_char(...TRUNCATED) | [{"attempts_by_match_type":{"infix":535,"latex_frac":1,"multi_op_chain":23,"nested_paren":2,"paren_e(...TRUNCATED) | [{"candidates_proposed":509,"candidates_validated":820,"new_span_finds":38,"novel_key_finds":29,"spa(...TRUNCATED) |
"Using ALL of the numbers 55, 36, 6, 3, 14, 99, 17, 82, create an expression that equals 245. You mu(...TRUNCATED) | "{\"numbers\": [55, 36, 6, 3, 14, 99, 17, 82], \"target\": 245, \"operators_used\": [\"*\", \"-\", \(...TRUNCATED) | countdown | [{"content":"You are a helpful assistant that solves math problems by reasoning step-by-step. Show y(...TRUNCATED) | [["Okay, let's see. I need to create an expression using all the numbers 55, 36, 6, 3, 14, 99, 17, 8(...TRUNCATED) | [
33991
] | [
4
] | [["Continue exploring alternative solution strategies.","Continue exploring alternative solution str(...TRUNCATED) | [["Okay, let's see. I need to create an expression using all the numbers 55, 36, 6, 3, 14, 99, 17, 8(...TRUNCATED) | iterative | [
"budget_reached"
] | [
null
] | false | [
0
] | [[["(82 + 99 + 55 + 14) - (17 - (36 / (6 - 3)))"],["(14 * 17) + (99 - 82) - (55 - 36 - 6 - 3)"],["(3(...TRUNCATED) | [
[
[
true
],
[
true
],
[
true
],
[
true
]
]
] | [[[{"answer_block":"(82 + 99 + 55 + 14) - (17 - (36 / (6 - 3)))","error":null,"final_answer":245.0,"(...TRUNCATED) | [
[]
] | [[{"all_operands":[99.0,2.0],"all_operators":["*"],"branch_id":0,"canonical_form":"99 * 2 = 198","ca(...TRUNCATED) | [[{"branch_id":0,"canonical_key":"expr:82=8","end_char":313,"end_char_in_round":313,"eval_value":"82(...TRUNCATED) | [[{"canonical_key":"subgoal:1","confidence":0.9,"end_char":820,"end_char_in_round":820,"goal_value":(...TRUNCATED) | [[{"end_char":1646,"end_char_in_round":1646,"marker_type":"not_helpful","round_idx":0,"span_text":"N(...TRUNCATED) | [[{"branch_id":0,"end_char":1101,"end_char_in_round":1101,"evidence_kind":"need","evidence_text":"ne(...TRUNCATED) | [[{"branch_id":0,"end_char":27502,"end_char_in_round":27502,"round_idx":0,"start_char":0,"start_char(...TRUNCATED) | [{"attempts_by_match_type":{"infix":802,"latex_frac":2,"multi_op_chain":85,"nested_paren":1,"paren_e(...TRUNCATED) | [{"candidates_proposed":683,"candidates_validated":1156,"new_span_finds":98,"novel_key_finds":54,"sp(...TRUNCATED) |
EAS-Qwen3-30B-A3B-Thinking-2507-cd8arg-iterative-stage02b-stage02a
Deterministic extraction of arithmetic attempts, expressions, subgoals, branches, and failure markers from Countdown reasoning traces
Dataset Info
- Rows: 3
- Columns: 26
Columns
| Column | Type | Description |
|---|---|---|
| question | Value('string') | No description provided |
| metadata | Value('string') | No description provided |
| task_source | Value('string') | No description provided |
| formatted_prompt | List({'content': Value('string'), 'role': Value('string')}) | No description provided |
| responses_by_sample | List(List(Value('string'))) | No description provided |
| tokens_by_sample | List(Value('int64')) | No description provided |
| num_rounds_by_sample | List(Value('int64')) | No description provided |
| continuation_prompts_by_sample | List(List(Value('string'))) | No description provided |
| continuation_text_by_sample | List(List(Value('string'))) | No description provided |
| strategy_type | Value('string') | No description provided |
| termination_reason_by_sample | List(Value('string')) | No description provided |
| correct_round_by_sample | List(Value('null')) | No description provided |
| smart_termination_enabled | Value('bool') | No description provided |
| retry_count_by_sample | List(Value('int64')) | No description provided |
| answers_by_sample | List(List(List(Value('string')))) | No description provided |
| evals_by_sample | List(List(List(Value('bool')))) | No description provided |
| eval_metadata_by_sample | List(List(List({'answer_block': Value('string'), 'error': Value('string'), 'final_answer': Value('float64'), 'is_correct': Value('bool')}))) | No description provided |
| annotations_by_sample | List(List(Value('null'))) | No description provided |
| arithmetic_attempts_by_sample | List(List({'all_operands': List(Value('float64')), 'all_operators': List(Value('string')), 'branch_id': Value('int64'), 'canonical_form': Value('string'), 'canonical_key': Value('string'), 'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'is_legal': Value('bool'), 'is_multi_operand': Value('bool'), 'is_valid_arithmetic': Value('bool'), 'match_type': Value('string'), 'operand_a': Value('float64'), 'operand_a_str': Value('string'), 'operand_b': Value('float64'), 'operand_b_str': Value('string'), 'operator': Value('string'), 'operator_raw': Value('string'), 'result': Value('float64'), 'result_str': Value('string'), 'round_idx': Value('int64'), 'separator_raw': Value('string'), 'span_text': Value('string'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64'), 'subgoal_id': Value('int64'), 'subgoal_value': Value('int64')})) | List of extracted binary arithmetic operations (A op B = C) per sample with spans and validation |
| expression_attempts_by_sample | List(List({'branch_id': Value('int64'), 'canonical_key': Value('string'), 'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'eval_value': Value('string'), 'expression': Value('string'), 'is_legal_initial': Value('bool'), 'is_valid_arithmetic': Value('bool'), 'numbers_used': List(Value('int64')), 'ops_used': List(Value('string')), 'relation': Value('string'), 'relation_raw': Value('string'), 'rhs': Value('string'), 'rhs_eval_value': Value('string'), 'round_idx': Value('int64'), 'source_kind': Value('string'), 'span_text': Value('string'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64'), 'subgoal_id': Value('int64'), 'subgoal_value': Value('int64')})) | List of extracted LHS = RHS expressions per sample with evaluation info |
| subgoals_by_sample | List(List({'canonical_key': Value('string'), 'confidence': Value('float64'), 'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'goal_value': Value('int64'), 'kind': Value('string'), 'round_idx': Value('int64'), 'span_text': Value('string'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64')})) | List of detected subgoal markers (goal, need, intermediate, eval_zero, make_zero) per sample with kind and confidence |
| failure_markers_by_sample | List(List({'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'marker_type': Value('string'), 'round_idx': Value('int64'), 'span_text': Value('string'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64')})) | List of detected failure/backtrack markers per sample |
| subgoal_windows_by_sample | List(List({'branch_id': Value('int64'), 'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'evidence_kind': Value('string'), 'evidence_text': Value('string'), 'round_idx': Value('int64'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64'), 'subgoal_id': Value('int64'), 'value': Value('int64')})) | List of subgoal window spans (where a subgoal is active) per sample |
| branches_by_sample | List(List({'branch_id': Value('int64'), 'end_char': Value('int64'), 'end_char_in_round': Value('int64'), 'round_idx': Value('int64'), 'start_char': Value('int64'), 'start_char_in_round': Value('int64'), 'start_kind': Value('string'), 'start_label': Value('string'), 'start_text': Value('string')})) | List of branch segments (strategy/alternative sections) per sample |
| extraction_stats_by_sample | List({'attempts_by_match_type': {'infix': Value('int64'), 'latex_frac': Value('int64'), 'multi_op_chain': Value('int64'), 'nested_paren': Value('int64'), 'paren_expr': Value('int64'), 'reversed': Value('int64')}, 'attempts_with_branch': Value('int64'), 'attempts_with_subgoal': Value('int64'), 'binary_attempts': Value('int64'), 'legal_expression_initial': Value('int64'), 'legal_initial': Value('int64'), 'merged_text_length': Value('int64'), 'multi_operand_attempts': Value('int64'), 'rounds_with_content': Value('int64'), 'subgoals_by_kind': {'eval_zero': Value('int64'), 'goal': Value('int64'), 'make_zero': Value('int64'), 'need': Value('int64'), 'need_more': Value('int64')}, 'total_attempts': Value('int64'), 'total_branches': Value('int64'), 'total_expression_attempts': Value('int64'), 'total_failure_markers': Value('int64'), 'total_subgoal_windows': Value('int64'), 'total_subgoals': Value('int64'), 'unique_canonical_keys': Value('int64'), 'unique_expression_keys': Value('int64'), 'valid_arithmetic': Value('int64'), 'valid_expression_arithmetic': Value('int64')}) | Summary statistics of extraction results per sample including subgoals_by_kind breakdown |
| llm_supplement_stats_by_sample | List({'candidates_proposed': Value('int64'), 'candidates_validated': Value('int64'), 'new_span_finds': Value('int64'), 'novel_key_finds': Value('int64'), 'span_duplicates': Value('int64'), 'total_attempts_after': Value('int64'), 'validation_failure_reasons': {'invalid_arithmetic': Value('int64'), 'no_parse': Value('int64'), 'not_found': Value('int64')}}) | No description provided |
Generation Parameters
{
"script_name": "stage_02a_extract_attempts.py",
"model": "deterministic_parser",
"hyperparameters": {},
"input_datasets": [
"reasoning-degeneration-dev/EAS-Qwen3-30B-A3B-Thinking-2507-cd8arg-iterative-stage02b"
],
"description": "Deterministic extraction of arithmetic attempts, expressions, subgoals, branches, and failure markers from Countdown reasoning traces"
}
Experiment Documentation
For complete experiment details, see https://github.com/reasoning-degeneration/ReasoningDegeneration/tree/main/experiments/degen_test_1_countdown/documentation
Usage
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/EAS-Qwen3-30B-A3B-Thinking-2507-cd8arg-iterative-stage02b-stage02a", split="train")
print(f"Loaded {len(dataset)} rows")
This dataset is tracked in reasoning-degeneration-dev/PROJECT-MANIFEST
- Downloads last month
- 10