Datasets:

---

SaifPunjwani

/

two-model-exploration-checkpoints

like 0

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
config: struct<accepted_selection_policy: string, actual_grpo_kl_coef: double, actual_grpo_lr: double, actua (... 1980 chars omitted)
  child 0, accepted_selection_policy: string
  child 1, actual_grpo_kl_coef: double
  child 2, actual_grpo_lr: double
  child 3, actual_online_grpo_max_steps: int64
  child 4, adam_beta1: double
  child 5, adam_beta2: double
  child 6, adam_epsilon: double
  child 7, benchmark_num_rollouts: int64
  child 8, benchmark_rollout_max_completion_len: int64
  child 9, correctness_zscore: string
  child 10, dataset_name: string
  child 11, distill_batch_size: int64
  child 12, distill_epochs: int64
  child 13, distill_lr: double
  child 14, dynamic_sampling: bool
  child 15, dynamic_sampling_max_attempts: int64
  child 16, enable_thinking: string
  child 17, format_penalty_weight: double
  child 18, gate_novelty_by_quality: bool
  child 19, grpo_advantage_normalization: string
  child 20, grpo_batch_size: int64
  child 21, grpo_clip_epsilon: double
  child 22, grpo_clip_epsilon_high: double
  child 23, grpo_grad_accum: int64
  child 24, grpo_kl_beta: double
  child 25, grpo_loss_aggregation: string
  child 26, grpo_loss_type: string
  child 27, grpo_lr: double
  child 28, grpo_max_steps: int64
  child 29, grpo_num_generations: int64
  child 30, grpo_updates_per_rollout: int64
  child 31, incorrect_novelty_scale: double
  child 32, init_checkpoint: string
  child 33, invalid_answer_penalty: double
  child 34, lambda_novelty: double
  child 35, logprob_c
...
uncated_completions: bool
  child 8, max_accepted_trajectories: int64
  child 9, max_distill_examples: int64
  child 10, min_accepted_trajectories: int64
  child 11, stage: string
  child 12, status: string
run_name: string
updated_at_unix: double
wandb_run_id: string
completion_token_length: int64
prompt_text: string
correctness_reward: double
invalid_answer_penalty: double
used_first_line: bool
blended_reward: double
format_penalty_breakdown: struct<final_answer_not_last_line: double, markdown_or_steps: double, too_many_lines: double, too_ma (... 17 chars omitted)
  child 0, final_answer_not_last_line: double
  child 1, markdown_or_steps: double
  child 2, too_many_lines: double
  child 3, too_many_words: double
base_policy: bool
ground_truth: string
r_novel_reward: double
novelty_reward: double
r_novel_raw: double
completion_text: string
format_penalty: double
novelty_text_contract: string
format_quality: double
r_correct_raw: double
finish_reason: string
is_correct: bool
r_novel_used: double
termination_penalty: double
reward: double
novelty_gate: double
novelty_text: string
snapshot_batch_index: int64
clipped: bool
terminated: bool
step: int64
r_correct_used: double
r_novel: double
valid_answer: bool
r_correct: double
r_novel_layers: struct<layer_7: double, layer_14: double, layer_21: double>
  child 0, layer_7: double
  child 1, layer_14: double
  child 2, layer_21: double
problem_id: string
completion_canonical: string
explorer_step: int64
format_penalty_weight: double
to
{'problem_id': Value('string'), 'prompt_text': Value('string'), 'completion_text': Value('string'), 'ground_truth': Value('string'), 'correctness_reward': Value('float64'), 'novelty_reward': Value('float64'), 'blended_reward': Value('float64'), 'reward': Value('float64'), 'r_correct': Value('float64'), 'r_novel': Value('float64'), 'is_correct': Value('bool'), 'explorer_step': Value('int64'), 'step': Value('int64'), 'finish_reason': Value('string'), 'completion_token_length': Value('int64'), 'clipped': Value('bool'), 'terminated': Value('bool'), 'valid_answer': Value('bool'), 'base_policy': Value('bool'), 'snapshot_batch_index': Value('int64'), 'r_correct_raw': Value('float64'), 'r_correct_used': Value('float64'), 'r_novel_raw': Value('float64'), 'r_novel_used': Value('float64'), 'r_novel_reward': Value('float64'), 'novelty_gate': Value('float64'), 'r_novel_layers': {'layer_7': Value('float64'), 'layer_14': Value('float64'), 'layer_21': Value('float64')}, 'termination_penalty': Value('float64'), 'invalid_answer_penalty': Value('float64'), 'format_quality': Value('float64'), 'format_penalty': Value('float64'), 'format_penalty_weight': Value('float64'), 'format_penalty_breakdown': {'final_answer_not_last_line': Value('float64'), 'markdown_or_steps': Value('float64'), 'too_many_lines': Value('float64'), 'too_many_words': Value('float64')}, 'novelty_text': Value('string'), 'novelty_text_contract': Value('string'), 'completion_canonical': Value('string'), 'used_first_line': Value('bool')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                         ^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/utils.py", line 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 295, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              config: struct<accepted_selection_policy: string, actual_grpo_kl_coef: double, actual_grpo_lr: double, actua (... 1980 chars omitted)
                child 0, accepted_selection_policy: string
                child 1, actual_grpo_kl_coef: double
                child 2, actual_grpo_lr: double
                child 3, actual_online_grpo_max_steps: int64
                child 4, adam_beta1: double
                child 5, adam_beta2: double
                child 6, adam_epsilon: double
                child 7, benchmark_num_rollouts: int64
                child 8, benchmark_rollout_max_completion_len: int64
                child 9, correctness_zscore: string
                child 10, dataset_name: string
                child 11, distill_batch_size: int64
                child 12, distill_epochs: int64
                child 13, distill_lr: double
                child 14, dynamic_sampling: bool
                child 15, dynamic_sampling_max_attempts: int64
                child 16, enable_thinking: string
                child 17, format_penalty_weight: double
                child 18, gate_novelty_by_quality: bool
                child 19, grpo_advantage_normalization: string
                child 20, grpo_batch_size: int64
                child 21, grpo_clip_epsilon: double
                child 22, grpo_clip_epsilon_high: double
                child 23, grpo_grad_accum: int64
                child 24, grpo_kl_beta: double
                child 25, grpo_loss_aggregation: string
                child 26, grpo_loss_type: string
                child 27, grpo_lr: double
                child 28, grpo_max_steps: int64
                child 29, grpo_num_generations: int64
                child 30, grpo_updates_per_rollout: int64
                child 31, incorrect_novelty_scale: double
                child 32, init_checkpoint: string
                child 33, invalid_answer_penalty: double
                child 34, lambda_novelty: double
                child 35, logprob_c
              ...
              uncated_completions: bool
                child 8, max_accepted_trajectories: int64
                child 9, max_distill_examples: int64
                child 10, min_accepted_trajectories: int64
                child 11, stage: string
                child 12, status: string
              run_name: string
              updated_at_unix: double
              wandb_run_id: string
              completion_token_length: int64
              prompt_text: string
              correctness_reward: double
              invalid_answer_penalty: double
              used_first_line: bool
              blended_reward: double
              format_penalty_breakdown: struct<final_answer_not_last_line: double, markdown_or_steps: double, too_many_lines: double, too_ma (... 17 chars omitted)
                child 0, final_answer_not_last_line: double
                child 1, markdown_or_steps: double
                child 2, too_many_lines: double
                child 3, too_many_words: double
              base_policy: bool
              ground_truth: string
              r_novel_reward: double
              novelty_reward: double
              r_novel_raw: double
              completion_text: string
              format_penalty: double
              novelty_text_contract: string
              format_quality: double
              r_correct_raw: double
              finish_reason: string
              is_correct: bool
              r_novel_used: double
              termination_penalty: double
              reward: double
              novelty_gate: double
              novelty_text: string
              snapshot_batch_index: int64
              clipped: bool
              terminated: bool
              step: int64
              r_correct_used: double
              r_novel: double
              valid_answer: bool
              r_correct: double
              r_novel_layers: struct<layer_7: double, layer_14: double, layer_21: double>
                child 0, layer_7: double
                child 1, layer_14: double
                child 2, layer_21: double
              problem_id: string
              completion_canonical: string
              explorer_step: int64
              format_penalty_weight: double
              to
              {'problem_id': Value('string'), 'prompt_text': Value('string'), 'completion_text': Value('string'), 'ground_truth': Value('string'), 'correctness_reward': Value('float64'), 'novelty_reward': Value('float64'), 'blended_reward': Value('float64'), 'reward': Value('float64'), 'r_correct': Value('float64'), 'r_novel': Value('float64'), 'is_correct': Value('bool'), 'explorer_step': Value('int64'), 'step': Value('int64'), 'finish_reason': Value('string'), 'completion_token_length': Value('int64'), 'clipped': Value('bool'), 'terminated': Value('bool'), 'valid_answer': Value('bool'), 'base_policy': Value('bool'), 'snapshot_batch_index': Value('int64'), 'r_correct_raw': Value('float64'), 'r_correct_used': Value('float64'), 'r_novel_raw': Value('float64'), 'r_novel_used': Value('float64'), 'r_novel_reward': Value('float64'), 'novelty_gate': Value('float64'), 'r_novel_layers': {'layer_7': Value('float64'), 'layer_14': Value('float64'), 'layer_21': Value('float64')}, 'termination_penalty': Value('float64'), 'invalid_answer_penalty': Value('float64'), 'format_quality': Value('float64'), 'format_penalty': Value('float64'), 'format_penalty_weight': Value('float64'), 'format_penalty_breakdown': {'final_answer_not_last_line': Value('float64'), 'markdown_or_steps': Value('float64'), 'too_many_lines': Value('float64'), 'too_many_words': Value('float64')}, 'novelty_text': Value('string'), 'novelty_text_contract': Value('string'), 'completion_canonical': Value('string'), 'used_first_line': Value('bool')}
              because column names don't match