Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
questions: list<item: struct<stage: string, question: string, options: list<item: string>, correct: int64, expl (... 17 chars omitted)
child 0, item: struct<stage: string, question: string, options: list<item: string>, correct: int64, explanation: st (... 5 chars omitted)
child 0, stage: string
child 1, question: string
child 2, options: list<item: string>
child 0, item: string
child 3, correct: int64
child 4, explanation: string
options: list<item: string>
child 0, item: string
explanation: string
question: string
stage: string
correct: int64
to
{'question': Value('string'), 'options': List(Value('string')), 'correct': Value('int64'), 'explanation': Value('string'), 'stage': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1779, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 299, 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
questions: list<item: struct<stage: string, question: string, options: list<item: string>, correct: int64, expl (... 17 chars omitted)
child 0, item: struct<stage: string, question: string, options: list<item: string>, correct: int64, explanation: st (... 5 chars omitted)
child 0, stage: string
child 1, question: string
child 2, options: list<item: string>
child 0, item: string
child 3, correct: int64
child 4, explanation: string
options: list<item: string>
child 0, item: string
explanation: string
question: string
stage: string
correct: int64
to
{'question': Value('string'), 'options': List(Value('string')), 'correct': Value('int64'), 'explanation': Value('string'), 'stage': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1832, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
question string | options list | correct int64 | explanation string | stage string |
|---|---|---|---|---|
What training objective does GPT use during pre-training? | [
"Masked language modeling (predicting masked tokens)",
"Next-token prediction: given previous tokens, predict the next one",
"Sentence classification",
"Image-text alignment"
] | 1 | GPT is a causal (autoregressive) language model trained with next-token prediction. Given tokens [t1, t2, ..., tn], it learns to predict tn+1. The loss is cross-entropy between predicted and actual next tokens. | pre |
How many transformer layers, attention heads, and embedding dimensions does GPT-2 Small (124M) have? | [
"6 layers, 6 heads, 512 dims",
"12 layers, 12 heads, 768 dims",
"24 layers, 16 heads, 1024 dims",
"48 layers, 25 heads, 1600 dims"
] | 1 | GPT-2 Small has 12 transformer layers, 12 attention heads per layer, and 768-dimensional embeddings. This architecture has 124 million parameters and can be trained on a single GPU in a few hours. | pre |
What is the role of the causal attention mask in GPT? | [
"It prevents attention to padding tokens",
"It prevents each token from attending to future tokens, ensuring the model can only use past context for predictions",
"It masks out low-confidence attention scores",
"It reduces memory usage during training"
] | 1 | The causal mask is a triangular matrix that sets future positions to -infinity before softmax. Token at position 5 can attend to positions 1-5 but not 6+. This ensures the model generates tokens left-to-right. | post |
What does 'temperature' control during text generation? | [
"The speed of generation",
"The randomness of token selection: lower temperature makes outputs more deterministic, higher makes them more diverse",
"The number of tokens generated",
"The model's confidence threshold"
] | 1 | Temperature divides logits before softmax. Temperature=0.1 makes the distribution very peaked (nearly deterministic). Temperature=1.0 is the training distribution. Temperature>1.0 flattens it, increasing randomness. | post |
Why does pre-training require significantly more compute than fine-tuning? | [
"Pre-training uses larger batch sizes",
"Pre-training processes trillions of tokens from scratch to learn general language patterns, while fine-tuning adjusts an already-capable model on thousands of examples",
"Pre-training uses a different architecture",
"Fine-tuning doesn't use gradients"
] | 1 | Pre-training builds all language knowledge from random weights over trillions of tokens. Fine-tuning starts from these learned weights and adjusts them on a much smaller dataset (thousands to millions of examples). | post |
README.md exists but content is empty.
- Downloads last month
- 10