Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 49 new columns ({'time_stats.attn_post_proj.median', 'time_stats.attn_pre_proj.mean', 'n_expanded_embd', 'time_stats.mlp_up_proj.median', 'time_stats.mlp_up_proj.std', 'time_stats.mlp_up_proj.min', 'time_stats.attn_post_proj.std', 'time_stats.input_norm_fused.median', 'time_stats.mlp_down_proj.median', 'time_stats.input_norm_fused.std', 'time_stats.mlp_down_proj.min', 'time_stats.attn_pre_proj.std', 'time_stats.attn_post_proj.max', 'time_stats.mlp_up_proj.max', 'time_stats.post_attention_norm_fused.max', 'time_stats.mlp_down_proj.std', 'time_stats.mlp_act.max', 'time_stats.mlp_act.median', 'time_stats.emb.median', 'time_stats.attn_rope.median', 'time_stats.attn_rope.std', 'time_stats.attn_pre_proj.max', 'time_stats.mlp_up_proj.mean', 'time_stats.attn_rope.max', 'time_stats.mlp_act.std', 'time_stats.post_attention_norm_fused.median', 'time_stats.mlp_act.mean', 'num_tokens', 'time_stats.attn_post_proj.mean', 'time_stats.emb.std', 'time_stats.mlp_act.min', 'time_stats.attn_pre_proj.min', 'time_stats.attn_pre_proj.median', 'time_stats.attn_post_proj.min', 'time_stats.input_norm_fused.max', 'time_stats.input_norm_fused.min', 'vocab_size', 'time_stats.post_attention_norm_fused.mean', 'time_stats.input_norm_fused.mean', 'time_stats.mlp_down_proj.max', 'time_stats.attn_rope.min', 'time_stats.attn_rope.mean', 'time_stats.emb.min', 'time_stats.emb.max', 'n_head', 'time_stats.post_attention_norm_fused.min', 'time_stats.mlp_down_proj.mean', 'time_stats.post_attention_norm_fused.std', 'time_stats.emb.mean'}) and 15 missing columns ({'attention_backend', 'max_model_len', 'prefill_chunk_size', 'time_stats.AttentionForward.max', 'time_stats.AttentionForward.std', 'time_stats.AttentionForward.mean', 'dtype', 'is_prefill', 'causal', 'time_stats.AttentionForward.median', 'batch_size', 'block_size', 'time_stats.AttentionForward.min', 'kv_cache_size', 'n_q_head'}).
This happened while the csv dataset builder was generating data using
/tmp/hf-datasets-cache/medium/datasets/12645371123672-config-parquet-and-info-project-vajra-dev-staging-e1f8aea7/hub/datasets--project-vajra--dev-staging-meta-llama-meta-llama-3-70b-instruct-h200-nvl/snapshots/edcd99ca727d6cabcc7fa7fef6e51a2ed2ccbe6c/mlp.csv.xz
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1831, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 714, in write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
time_stats.mlp_up_proj.min: double
time_stats.mlp_up_proj.max: double
time_stats.mlp_up_proj.mean: double
time_stats.mlp_up_proj.median: double
time_stats.mlp_up_proj.std: double
time_stats.post_attention_norm_fused.min: double
time_stats.post_attention_norm_fused.max: double
time_stats.post_attention_norm_fused.mean: double
time_stats.post_attention_norm_fused.median: double
time_stats.post_attention_norm_fused.std: double
time_stats.attn_post_proj.min: double
time_stats.attn_post_proj.max: double
time_stats.attn_post_proj.mean: double
time_stats.attn_post_proj.median: double
time_stats.attn_post_proj.std: double
time_stats.mlp_down_proj.min: double
time_stats.mlp_down_proj.max: double
time_stats.mlp_down_proj.mean: double
time_stats.mlp_down_proj.median: double
time_stats.mlp_down_proj.std: double
time_stats.attn_rope.min: double
time_stats.attn_rope.max: double
time_stats.attn_rope.mean: double
time_stats.attn_rope.median: double
time_stats.attn_rope.std: double
time_stats.attn_pre_proj.min: double
time_stats.attn_pre_proj.max: double
time_stats.attn_pre_proj.mean: double
time_stats.attn_pre_proj.median: double
time_stats.attn_pre_proj.std: double
time_stats.input_norm_fused.min: double
time_stats.input_norm_fused.max: double
time_stats.input_norm_fused.mean: double
time_stats.input_norm_fused.median: double
time_stats.input_norm_fused.std: double
time_stats.mlp_act.min: double
time_stats.mlp_act.max: double
time_stats.mlp_act.mean: double
time_stats.mlp_act.median: double
time_stats.mlp_act.std: double
time_stats.emb.min: double
time_stats.emb.max: double
time_stats.emb.mean: double
time_stats.emb.median: double
time_stats.emb.std: double
n_head: int64
n_kv_head: int64
n_embd: int64
n_expanded_embd: int64
vocab_size: int64
num_tokens: int64
num_tensor_parallel_workers: int64
-- schema metadata --
pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 8184
to
{'time_stats.AttentionForward.min': Value('float64'), 'time_stats.AttentionForward.max': Value('float64'), 'time_stats.AttentionForward.mean': Value('float64'), 'time_stats.AttentionForward.median': Value('float64'), 'time_stats.AttentionForward.std': Value('float64'), 'n_embd': Value('int64'), 'n_q_head': Value('int64'), 'n_kv_head': Value('int64'), 'block_size': Value('int64'), 'num_tensor_parallel_workers': Value('int64'), 'max_model_len': Value('int64'), 'batch_size': Value('int64'), 'prefill_chunk_size': Value('int64'), 'kv_cache_size': Value('int64'), 'is_prefill': Value('bool'), 'attention_backend': Value('string'), 'dtype': Value('string'), 'causal': Value('bool')}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1339, 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 972, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 894, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 970, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1702, 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 1833, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 49 new columns ({'time_stats.attn_post_proj.median', 'time_stats.attn_pre_proj.mean', 'n_expanded_embd', 'time_stats.mlp_up_proj.median', 'time_stats.mlp_up_proj.std', 'time_stats.mlp_up_proj.min', 'time_stats.attn_post_proj.std', 'time_stats.input_norm_fused.median', 'time_stats.mlp_down_proj.median', 'time_stats.input_norm_fused.std', 'time_stats.mlp_down_proj.min', 'time_stats.attn_pre_proj.std', 'time_stats.attn_post_proj.max', 'time_stats.mlp_up_proj.max', 'time_stats.post_attention_norm_fused.max', 'time_stats.mlp_down_proj.std', 'time_stats.mlp_act.max', 'time_stats.mlp_act.median', 'time_stats.emb.median', 'time_stats.attn_rope.median', 'time_stats.attn_rope.std', 'time_stats.attn_pre_proj.max', 'time_stats.mlp_up_proj.mean', 'time_stats.attn_rope.max', 'time_stats.mlp_act.std', 'time_stats.post_attention_norm_fused.median', 'time_stats.mlp_act.mean', 'num_tokens', 'time_stats.attn_post_proj.mean', 'time_stats.emb.std', 'time_stats.mlp_act.min', 'time_stats.attn_pre_proj.min', 'time_stats.attn_pre_proj.median', 'time_stats.attn_post_proj.min', 'time_stats.input_norm_fused.max', 'time_stats.input_norm_fused.min', 'vocab_size', 'time_stats.post_attention_norm_fused.mean', 'time_stats.input_norm_fused.mean', 'time_stats.mlp_down_proj.max', 'time_stats.attn_rope.min', 'time_stats.attn_rope.mean', 'time_stats.emb.min', 'time_stats.emb.max', 'n_head', 'time_stats.post_attention_norm_fused.min', 'time_stats.mlp_down_proj.mean', 'time_stats.post_attention_norm_fused.std', 'time_stats.emb.mean'}) and 15 missing columns ({'attention_backend', 'max_model_len', 'prefill_chunk_size', 'time_stats.AttentionForward.max', 'time_stats.AttentionForward.std', 'time_stats.AttentionForward.mean', 'dtype', 'is_prefill', 'causal', 'time_stats.AttentionForward.median', 'batch_size', 'block_size', 'time_stats.AttentionForward.min', 'kv_cache_size', 'n_q_head'}).
This happened while the csv dataset builder was generating data using
/tmp/hf-datasets-cache/medium/datasets/12645371123672-config-parquet-and-info-project-vajra-dev-staging-e1f8aea7/hub/datasets--project-vajra--dev-staging-meta-llama-meta-llama-3-70b-instruct-h200-nvl/snapshots/edcd99ca727d6cabcc7fa7fef6e51a2ed2ccbe6c/mlp.csv.xz
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)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.
time_stats.AttentionForward.min float64 | time_stats.AttentionForward.max float64 | time_stats.AttentionForward.mean float64 | time_stats.AttentionForward.median float64 | time_stats.AttentionForward.std float64 | n_embd int64 | n_q_head int64 | n_kv_head int64 | block_size int64 | num_tensor_parallel_workers int64 | max_model_len int64 | batch_size int64 | prefill_chunk_size int64 | kv_cache_size int64 | is_prefill bool | attention_backend string | dtype string | causal bool |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.085763 | 0.085763 | 0.085763 | 0.085763 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 0 | true | flashinfer_auto | torch.bfloat16 | true |
0.067331 | 0.067331 | 0.067331 | 0.067331 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 32 | true | flashinfer_auto | torch.bfloat16 | true |
0.10202 | 0.10202 | 0.10202 | 0.10202 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 64 | true | flashinfer_auto | torch.bfloat16 | true |
0.078466 | 0.078466 | 0.078466 | 0.078466 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 96 | true | flashinfer_auto | torch.bfloat16 | true |
0.07725 | 0.07725 | 0.07725 | 0.07725 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 128 | true | flashinfer_auto | torch.bfloat16 | true |
0.073539 | 0.073539 | 0.073539 | 0.073539 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 160 | true | flashinfer_auto | torch.bfloat16 | true |
0.077699 | 0.077699 | 0.077699 | 0.077699 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 192 | true | flashinfer_auto | torch.bfloat16 | true |
0.083651 | 0.083651 | 0.083651 | 0.083651 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 224 | true | flashinfer_auto | torch.bfloat16 | true |
0.070979 | 0.070979 | 0.070979 | 0.070979 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 256 | true | flashinfer_auto | torch.bfloat16 | true |
0.075267 | 0.075267 | 0.075267 | 0.075267 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 288 | true | flashinfer_auto | torch.bfloat16 | true |
0.071875 | 0.071875 | 0.071875 | 0.071875 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 320 | true | flashinfer_auto | torch.bfloat16 | true |
0.086051 | 0.086051 | 0.086051 | 0.086051 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 352 | true | flashinfer_auto | torch.bfloat16 | true |
0.079267 | 0.079267 | 0.079267 | 0.079267 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 384 | true | flashinfer_auto | torch.bfloat16 | true |
0.077348 | 0.077348 | 0.077348 | 0.077348 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 416 | true | flashinfer_auto | torch.bfloat16 | true |
0.079107 | 0.079107 | 0.079107 | 0.079107 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 448 | true | flashinfer_auto | torch.bfloat16 | true |
0.084931 | 0.084931 | 0.084931 | 0.084931 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 480 | true | flashinfer_auto | torch.bfloat16 | true |
0.079107 | 0.079107 | 0.079107 | 0.079107 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 512 | true | flashinfer_auto | torch.bfloat16 | true |
0.089443 | 0.089443 | 0.089443 | 0.089443 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 544 | true | flashinfer_auto | torch.bfloat16 | true |
0.080227 | 0.080227 | 0.080227 | 0.080227 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 576 | true | flashinfer_auto | torch.bfloat16 | true |
0.088291 | 0.088291 | 0.088291 | 0.088291 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 608 | true | flashinfer_auto | torch.bfloat16 | true |
0.082979 | 0.082979 | 0.082979 | 0.082979 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 640 | true | flashinfer_auto | torch.bfloat16 | true |
0.084995 | 0.084995 | 0.084995 | 0.084995 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 672 | true | flashinfer_auto | torch.bfloat16 | true |
0.080451 | 0.080451 | 0.080451 | 0.080451 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 704 | true | flashinfer_auto | torch.bfloat16 | true |
0.086819 | 0.086819 | 0.086819 | 0.086819 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 736 | true | flashinfer_auto | torch.bfloat16 | true |
0.083139 | 0.083139 | 0.083139 | 0.083139 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 768 | true | flashinfer_auto | torch.bfloat16 | true |
0.088163 | 0.088163 | 0.088163 | 0.088163 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 800 | true | flashinfer_auto | torch.bfloat16 | true |
0.08522 | 0.08522 | 0.08522 | 0.08522 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 832 | true | flashinfer_auto | torch.bfloat16 | true |
0.09242 | 0.09242 | 0.09242 | 0.09242 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 864 | true | flashinfer_auto | torch.bfloat16 | true |
0.087682 | 0.087682 | 0.087682 | 0.087682 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 896 | true | flashinfer_auto | torch.bfloat16 | true |
0.095107 | 0.095107 | 0.095107 | 0.095107 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 928 | true | flashinfer_auto | torch.bfloat16 | true |
0.084995 | 0.084995 | 0.084995 | 0.084995 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 960 | true | flashinfer_auto | torch.bfloat16 | true |
0.093602 | 0.093602 | 0.093602 | 0.093602 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 992 | true | flashinfer_auto | torch.bfloat16 | true |
0.103171 | 0.103171 | 0.103171 | 0.103171 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,024 | true | flashinfer_auto | torch.bfloat16 | true |
0.088162 | 0.088162 | 0.088162 | 0.088162 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,056 | true | flashinfer_auto | torch.bfloat16 | true |
0.089923 | 0.089923 | 0.089923 | 0.089923 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,088 | true | flashinfer_auto | torch.bfloat16 | true |
0.092707 | 0.092707 | 0.092707 | 0.092707 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,120 | true | flashinfer_auto | torch.bfloat16 | true |
0.109091 | 0.109091 | 0.109091 | 0.109091 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,152 | true | flashinfer_auto | torch.bfloat16 | true |
0.101475 | 0.101475 | 0.101475 | 0.101475 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,184 | true | flashinfer_auto | torch.bfloat16 | true |
0.093891 | 0.093891 | 0.093891 | 0.093891 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,216 | true | flashinfer_auto | torch.bfloat16 | true |
0.097476 | 0.097476 | 0.097476 | 0.097476 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,248 | true | flashinfer_auto | torch.bfloat16 | true |
0.115139 | 0.115139 | 0.115139 | 0.115139 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,280 | true | flashinfer_auto | torch.bfloat16 | true |
0.09533 | 0.09533 | 0.09533 | 0.09533 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,312 | true | flashinfer_auto | torch.bfloat16 | true |
0.094466 | 0.094466 | 0.094466 | 0.094466 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,344 | true | flashinfer_auto | torch.bfloat16 | true |
0.098467 | 0.098467 | 0.098467 | 0.098467 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,376 | true | flashinfer_auto | torch.bfloat16 | true |
0.107396 | 0.107396 | 0.107396 | 0.107396 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,408 | true | flashinfer_auto | torch.bfloat16 | true |
0.098179 | 0.098179 | 0.098179 | 0.098179 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,440 | true | flashinfer_auto | torch.bfloat16 | true |
0.096803 | 0.096803 | 0.096803 | 0.096803 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,472 | true | flashinfer_auto | torch.bfloat16 | true |
0.100867 | 0.100867 | 0.100867 | 0.100867 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,504 | true | flashinfer_auto | torch.bfloat16 | true |
0.088131 | 0.088131 | 0.088131 | 0.088131 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,536 | true | flashinfer_auto | torch.bfloat16 | true |
0.099557 | 0.099557 | 0.099557 | 0.099557 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,568 | true | flashinfer_auto | torch.bfloat16 | true |
0.088068 | 0.088068 | 0.088068 | 0.088068 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,600 | true | flashinfer_auto | torch.bfloat16 | true |
0.091363 | 0.091363 | 0.091363 | 0.091363 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,632 | true | flashinfer_auto | torch.bfloat16 | true |
0.091939 | 0.091939 | 0.091939 | 0.091939 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,664 | true | flashinfer_auto | torch.bfloat16 | true |
0.090115 | 0.090115 | 0.090115 | 0.090115 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,696 | true | flashinfer_auto | torch.bfloat16 | true |
0.091431 | 0.091431 | 0.091431 | 0.091431 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,728 | true | flashinfer_auto | torch.bfloat16 | true |
0.091875 | 0.091875 | 0.091875 | 0.091875 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,760 | true | flashinfer_auto | torch.bfloat16 | true |
0.091907 | 0.091907 | 0.091907 | 0.091907 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,792 | true | flashinfer_auto | torch.bfloat16 | true |
0.096898 | 0.096898 | 0.096898 | 0.096898 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,824 | true | flashinfer_auto | torch.bfloat16 | true |
0.058689 | 0.058689 | 0.058689 | 0.058689 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,856 | true | flashinfer_auto | torch.bfloat16 | true |
0.092834 | 0.092834 | 0.092834 | 0.092834 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,888 | true | flashinfer_auto | torch.bfloat16 | true |
0.096547 | 0.096547 | 0.096547 | 0.096547 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,920 | true | flashinfer_auto | torch.bfloat16 | true |
0.093572 | 0.093572 | 0.093572 | 0.093572 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,952 | true | flashinfer_auto | torch.bfloat16 | true |
0.060737 | 0.060737 | 0.060737 | 0.060737 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 1,984 | true | flashinfer_auto | torch.bfloat16 | true |
0.098052 | 0.098052 | 0.098052 | 0.098052 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,016 | true | flashinfer_auto | torch.bfloat16 | true |
0.097187 | 0.097187 | 0.097187 | 0.097187 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,048 | true | flashinfer_auto | torch.bfloat16 | true |
0.095587 | 0.095587 | 0.095587 | 0.095587 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,080 | true | flashinfer_auto | torch.bfloat16 | true |
0.073475 | 0.073475 | 0.073475 | 0.073475 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,112 | true | flashinfer_auto | torch.bfloat16 | true |
0.113987 | 0.113987 | 0.113987 | 0.113987 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,144 | true | flashinfer_auto | torch.bfloat16 | true |
0.097509 | 0.097509 | 0.097509 | 0.097509 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,176 | true | flashinfer_auto | torch.bfloat16 | true |
0.100066 | 0.100066 | 0.100066 | 0.100066 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,208 | true | flashinfer_auto | torch.bfloat16 | true |
0.064547 | 0.064547 | 0.064547 | 0.064547 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,240 | true | flashinfer_auto | torch.bfloat16 | true |
0.101859 | 0.101859 | 0.101859 | 0.101859 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,272 | true | flashinfer_auto | torch.bfloat16 | true |
0.102243 | 0.102243 | 0.102243 | 0.102243 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,304 | true | flashinfer_auto | torch.bfloat16 | true |
0.100164 | 0.100164 | 0.100164 | 0.100164 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,336 | true | flashinfer_auto | torch.bfloat16 | true |
0.066308 | 0.066308 | 0.066308 | 0.066308 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,368 | true | flashinfer_auto | torch.bfloat16 | true |
0.105093 | 0.105093 | 0.105093 | 0.105093 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,400 | true | flashinfer_auto | torch.bfloat16 | true |
0.102818 | 0.102818 | 0.102818 | 0.102818 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,432 | true | flashinfer_auto | torch.bfloat16 | true |
0.101156 | 0.101156 | 0.101156 | 0.101156 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,464 | true | flashinfer_auto | torch.bfloat16 | true |
0.069026 | 0.069026 | 0.069026 | 0.069026 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,496 | true | flashinfer_auto | torch.bfloat16 | true |
0.105284 | 0.105284 | 0.105284 | 0.105284 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,528 | true | flashinfer_auto | torch.bfloat16 | true |
0.111331 | 0.111331 | 0.111331 | 0.111331 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,560 | true | flashinfer_auto | torch.bfloat16 | true |
0.128292 | 0.128292 | 0.128292 | 0.128292 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,592 | true | flashinfer_auto | torch.bfloat16 | true |
0.073186 | 0.073186 | 0.073186 | 0.073186 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,624 | true | flashinfer_auto | torch.bfloat16 | true |
0.115298 | 0.115298 | 0.115298 | 0.115298 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,656 | true | flashinfer_auto | torch.bfloat16 | true |
0.116292 | 0.116292 | 0.116292 | 0.116292 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,688 | true | flashinfer_auto | torch.bfloat16 | true |
0.113828 | 0.113828 | 0.113828 | 0.113828 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,720 | true | flashinfer_auto | torch.bfloat16 | true |
0.076995 | 0.076995 | 0.076995 | 0.076995 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,752 | true | flashinfer_auto | torch.bfloat16 | true |
0.122051 | 0.122051 | 0.122051 | 0.122051 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,784 | true | flashinfer_auto | torch.bfloat16 | true |
0.109507 | 0.109507 | 0.109507 | 0.109507 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,816 | true | flashinfer_auto | torch.bfloat16 | true |
0.111075 | 0.111075 | 0.111075 | 0.111075 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,848 | true | flashinfer_auto | torch.bfloat16 | true |
0.074689 | 0.074689 | 0.074689 | 0.074689 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,880 | true | flashinfer_auto | torch.bfloat16 | true |
0.109571 | 0.109571 | 0.109571 | 0.109571 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,912 | true | flashinfer_auto | torch.bfloat16 | true |
0.108388 | 0.108388 | 0.108388 | 0.108388 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,944 | true | flashinfer_auto | torch.bfloat16 | true |
0.110883 | 0.110883 | 0.110883 | 0.110883 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 2,976 | true | flashinfer_auto | torch.bfloat16 | true |
0.074851 | 0.074851 | 0.074851 | 0.074851 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,008 | true | flashinfer_auto | torch.bfloat16 | true |
0.111779 | 0.111779 | 0.111779 | 0.111779 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,040 | true | flashinfer_auto | torch.bfloat16 | true |
0.112161 | 0.112161 | 0.112161 | 0.112161 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,072 | true | flashinfer_auto | torch.bfloat16 | true |
0.111427 | 0.111427 | 0.111427 | 0.111427 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,104 | true | flashinfer_auto | torch.bfloat16 | true |
0.078627 | 0.078627 | 0.078627 | 0.078627 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,136 | true | flashinfer_auto | torch.bfloat16 | true |
0.116132 | 0.116132 | 0.116132 | 0.116132 | 0 | 8,192 | 64 | 8 | 16 | 1 | 8,192 | 1 | 32 | 3,168 | true | flashinfer_auto | torch.bfloat16 | true |
End of preview.
No dataset card yet
- Downloads last month
- 3