Hugging Face's logo

Datasets:
Arushhh
/
dlm-exp1-drafter-voting-research-artifacts

License:
Dataset card Files
xet
 Community
Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
acc: double
ce: double
cert: double
grad_norm: double
loss: double
n_correct: double
n_rows: double
top1: double
uniq_experts: double
step: int64
lr: double
pool_M: int64
tokens_per_s: double
sec_per_step: double
utc: timestamp[s]
allmask_pass1_acc: double
allmask_pass1_conf_frac: double
allmask_pass1_conf_precision: double
allmask_pass1_states: double
allmask_pass1_top1: double
draft_pass1_acc: double
draft_pass1_conf_frac: double
draft_pass1_conf_precision: double
draft_pass1_states: double
draft_pass1_top1: double
event: string
config_hash: string
r: int64
model: struct<hidden_size: int64, num_experts: int64, num_experts_per_tok: int64, num_hidden_layers: int64, (... 19 chars omitted)
  child 0, hidden_size: int64
  child 1, num_experts: int64
  child 2, num_experts_per_tok: int64
  child 3, num_hidden_layers: int64
  child 4, vocab_size: int64
alpha: double
train_gates: bool
dropout: double
targets: list<item: string>
  child 0, item: string
n_tensors: int64
to
{'alpha': Value('float64'), 'config_hash': Value('string'), 'dropout': Value('float64'), 'model': {'hidden_size': Value('int64'), 'num_experts': Value('int64'), 'num_experts_per_tok': Value('int64'), 'num_hidden_layers': Value('int64'), 'vocab_size': Value('int64')}, 'n_tensors': Value('int64'), 'r': Value('int64'), 'targets': List(Value('string')), 'train_gates': Value('bool')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                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 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              acc: double
              ce: double
              cert: double
              grad_norm: double
              loss: double
              n_correct: double
              n_rows: double
              top1: double
              uniq_experts: double
              step: int64
              lr: double
              pool_M: int64
              tokens_per_s: double
              sec_per_step: double
              utc: timestamp[s]
              allmask_pass1_acc: double
              allmask_pass1_conf_frac: double
              allmask_pass1_conf_precision: double
              allmask_pass1_states: double
              allmask_pass1_top1: double
              draft_pass1_acc: double
              draft_pass1_conf_frac: double
              draft_pass1_conf_precision: double
              draft_pass1_states: double
              draft_pass1_top1: double
              event: string
              config_hash: string
              r: int64
              model: struct<hidden_size: int64, num_experts: int64, num_experts_per_tok: int64, num_hidden_layers: int64, (... 19 chars omitted)
                child 0, hidden_size: int64
                child 1, num_experts: int64
                child 2, num_experts_per_tok: int64
                child 3, num_hidden_layers: int64
                child 4, vocab_size: int64
              alpha: double
              train_gates: bool
              dropout: double
              targets: list<item: string>
                child 0, item: string
              n_tensors: int64
              to
              {'alpha': Value('float64'), 'config_hash': Value('string'), 'dropout': Value('float64'), 'model': {'hidden_size': Value('int64'), 'num_experts': Value('int64'), 'num_experts_per_tok': Value('int64'), 'num_hidden_layers': Value('int64'), 'vocab_size': Value('int64')}, 'n_tensors': Value('int64'), 'r': Value('int64'), 'targets': List(Value('string')), 'train_gates': Value('bool')}
              because column names don't match

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

dlm-exp1 drafter-voting: research artifacts (public mirror)

Everything bulky from the drafter-voting branch of https://github.com/NoviceCoderInfinity/dlm-general-expert-exp1 that git does not track (uploaded 2026-09-06 before the compute instance was destroyed):

  • logs/research_v1/: baseline cells (one JSON line per item; per-pass NPZ traces for both checkpoints in baselines_traces/ and baselines_b32_traces/), state-batch fidelity probe states (fidelity/), flush-fusion boundaries, console logs and the launch log.
  • logs/desvote/, logs/desvote_traces/: the earlier DES-Vote 500-item cells and expert-set traces (block-4 model at block 32).
  • results/: metrics, audits and reports (also on the branch). docs/: handoffs, protocol, Claude session transcripts (docs/claude_sessions/, tokens masked). code/: snapshot at upload time (the git branch is canonical).
  • codex_session.md: the Codex supervision chat.

Models are not re-uploaded: block-32 checkpoint JetLM/SDAR-30B-A3B-Chat-b32 (revision c351bbc37d240aa6871f167e8f92d694281b0c22), block-4 checkpoint JetLM/SDAR-30B-A3B-Chat (revision f5add2a159163a2a8f07e9da7dfcbfaadc73d6d4). The friend's Stage-0 routing traces are in Anupam-Rawat-IITB/stage0-routing-logs.

Start with results/reports/RESEARCH_RECORD_2026-09-05.md (every hypothesis and experiment, with verdicts) and results/reports/research_v1_decision.md.
Downloads last month
275
Total file size:
138 GB