Skip to content

Commit 3310f19

Browse files
committed
Fix spacing issues
1 parent 44764e3 commit 3310f19

File tree

3 files changed

+3
-199
lines changed

3 files changed

+3
-199
lines changed

examples/glue/README.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ pip install -r requirements.txt
99

1010
## Training
1111

12-
To fine-tune BERT on a specific dataset, run the following command (we are using the `SST2` dataset in this example):
12+
To fine-tune BERT on a specific dataset, run the following command (we are using the SST2 dataset in this example):
1313

1414
```bash
1515
python train.py --dataset_name sst2 \
@@ -36,7 +36,7 @@ python analyze.py --dataset_name sst2 \
3636
--factor_strategy ekfac
3737
```
3838

39-
On an A100 (80GB), it takes roughly 95 minutes to compute the pairwise scores for `SST2` (including computing EKFAC factors):
39+
On an A100 (80GB), it takes roughly 90 minutes to compute the pairwise scores for SST2 (including computing EKFAC factors):
4040

4141
```
4242
----------------------------------------------------------------------------------------------------------------------------------

examples/openwebtext/analyze.py

Lines changed: 0 additions & 196 deletions
This file was deleted.

examples/swag/README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -36,7 +36,7 @@ python analyze.py --factor_batch_size 128 \
3636
```
3737

3838
Alternative options for `factor_strategy` include `identity`, `diagonal`, and `kfac`.
39-
On an A100 (80GB), computing the pairwise scores (including EKFAC factors) takes approximately 3 hours:
39+
On an A100 (80GB), computing the pairwise scores (including EKFAC factors) takes approximately 10 hours:
4040

4141
```
4242
----------------------------------------------------------------------------------------------------------------------------------

0 commit comments

Comments
 (0)