 Sebastian Raschka
|
c21bfe4a23
Add PyPI package (#576)
|
8 月之前 |
|
Sebastian Raschka
|
a08d7aaa84
Uv workflow improvements (#531)
|
9 月之前 |
|
Sebastian Raschka
|
68e2efe1c9
Mention small discrepancy due to Dropout non-reproducibility in PyTorch (#519)
|
9 月之前 |
|
rasbt
|
dc1b1a05b0
note about random numbers
|
1 年之前 |
|
Sebastian Raschka
|
222f7b16f8
update gpt-2 paper url
|
1 年之前 |
|
rasbt
|
8ad50a3315
update gpt-2 paper link
|
1 年之前 |
|
rasbt
|
1e48c13e89
update gpt-2 paper link
|
1 年之前 |
|
Sebastian Raschka
|
08040f024c
Test code in pytorch 2.4 (#285)
|
1 年之前 |
 Thanh Tran
|
070a69fc8b
fix typos & inconsistent texts (#269)
|
1 年之前 |
 Jeroen Van Goey
|
48bd72c890
fix typos, add codespell pre-commit hook (#264)
|
1 年之前 |
|
rasbt
|
6ffd628bb6
add missing "be" to figure
|
1 年之前 |
|
rasbt
|
921e91a05f
use correct chapter reference
|
1 年之前 |
|
rasbt
|
31806828d0
add links to summary sections
|
1 年之前 |
|
rasbt
|
796f0e2a30
add clarifying note about GELU
|
1 年之前 |
|
rasbt
|
ab23ca5b1b
force refresh figure
|
1 年之前 |
|
rasbt
|
6a8acf5135
remove redundant plus sign
|
1 年之前 |
 Daniel Kleine
|
81c843bdc0
minor fixes (#246)
|
1 年之前 |
|
rasbt
|
98d453b666
update formatting
|
1 年之前 |
|
rasbt
|
e5e6aaf9f1
flops analysis
|
1 年之前 |
|
rasbt
|
c735c21e87
fix swiglu acronym
|
1 年之前 |
|
Sebastian Raschka
|
97ed38116a
Rename drop_resid to drop_shortcut (#136)
|
1 年之前 |
|
rasbt
|
d202cabdee
update figures
|
1 年之前 |
|
rasbt
|
6de0417321
cleanup
|
1 年之前 |
|
Sebastian Raschka
|
2de60d1bfb
Rename variable to context_length to make it easier on readers (#106)
|
1 年之前 |
|
Sebastian Raschka
|
3829ccdb34
Remove reundant dropout in MLP module (#105)
|
1 年之前 |
|
Sebastian Raschka
|
a2cd8436cb
Ch05 supplementary code (#81)
|
1 年之前 |
|
rasbt
|
4fc6de7afa
add notes
|
1 年之前 |
|
rasbt
|
d60da19fd0
add more notes and embed figures externally to save space
|
1 年之前 |
|
rasbt
|
861c296312
add imports and version on top
|
1 年之前 |
 joel-foo
|
dbb5e65a29
Remove duplicate cells
|
1 年之前 |