Sebastian Raschka
3654571184
align formulas in notes with code ( #605 )
2025-04-06 16:46:53 -05:00
Greg Gandenberger
b92c0dff89
Add note about context_length ( #549 )
...
* Add note about context_length
* update note
---------
Co-authored-by: rasbt <mail@sebastianraschka.com>
2025-02-27 08:36:41 -06:00
Sebastian Raschka
a08d7aaa84
Uv workflow improvements ( #531 )
...
* Uv workflow improvements
* Uv workflow improvements
* linter improvements
* pytproject.toml fixes
* pytproject.toml fixes
* pytproject.toml fixes
* pytproject.toml fixes
* pytproject.toml fixes
* pytproject.toml fixes
* windows fixes
* windows fixes
* windows fixes
* windows fixes
* windows fixes
* windows fixes
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
* win32 fix
2025-02-16 13:16:51 -06:00
rasbt
1183fd7837
add dropout scaling note
2024-11-06 05:52:47 -06:00
Sebastian Raschka
08040f024c
Test code in pytorch 2.4 ( #285 )
...
* test code in pytorch 2.4
* update
2024-07-24 21:53:41 -05:00
rasbt
31806828d0
add links to summary sections
2024-06-29 07:33:26 -05:00
rasbt
c7f892550e
add clarification about :num_tokens
2024-06-29 07:16:42 -05:00
rasbt
283397aaf2
add main and optional sections
2024-06-19 17:48:25 -05:00
rasbt
5d1fbbbfd2
update dotted line
2024-06-17 20:17:56 -05:00
rasbt
aaa54b10b3
dim-consistency
2024-06-12 19:43:25 -05:00
Sebastian Raschka
72a073bbbf
Remove leftover instances of self.tokenizer ( #201 )
...
* Remove leftover instances of self.tokenizer
* add endoftext token
2024-06-08 14:57:34 -05:00
Sebastian Raschka
c303a7f36d
Explain value truncation in some figures ( #199 )
...
* clarify truncation
* typo fix
2024-06-08 13:24:37 -05:00
rasbt
1e12da90e6
clarify truncation
2024-06-08 13:13:43 -05:00
rasbt
42af52fef4
revert unnecessary changes
2024-05-27 07:37:06 -05:00
rasbt
87c3e78dcb
Revert "Revert "newline""
...
This reverts commit a53ca10508aacc0c2a8da5467626f5fce33ef162.
2024-05-27 07:32:45 -05:00
rasbt
a53ca10508
Revert "newline"
...
This reverts commit 23982ed3fabc277ca50a998c325d304940eb78a5.
2024-05-27 07:32:22 -05:00
rasbt
23982ed3fa
newline
2024-05-27 07:30:27 -05:00
rasbt
050c8b7b73
update pr
2024-05-26 15:38:35 -05:00
Kostyantyn Borysenko
76cdf5e299
Fix an incorrect input dimension
2024-05-26 13:05:07 -07:00
rasbt
98d453b666
update formatting
2024-05-24 07:20:37 -05:00
rasbt
3b57b6d8c4
make consistent with the latest production version
2024-05-18 12:08:39 -05:00
Sebastian Raschka
7740d556a0
Use dim=-1 for consistency ( #122 )
2024-04-18 05:56:23 -05:00
James Holcombe
05718c6b94
Use instance tokenizer ( #116 )
...
* Use instance tokenizer
* consistency updates
---------
Co-authored-by: Sebastian Raschka <mail@sebastianraschka.com>
2024-04-10 21:16:19 -04:00
Sebastian Raschka
2de60d1bfb
Rename variable to context_length to make it easier on readers ( #106 )
...
* rename to context length
* fix spacing
2024-04-04 07:27:41 -05:00
rasbt
7d1eadd0be
update notes
2024-04-02 18:27:13 -05:00
Intelligence-Manifesto
96b1fde3f1
"Typographical error ( #104 )
2024-04-02 18:07:21 -05:00
rasbt
3ad442ee90
skip version cell
2024-03-28 08:23:33 -05:00
Sebastian Raschka
a2cd8436cb
Ch05 supplementary code ( #81 )
2024-03-19 09:26:26 -05:00
Sebastian Raschka
ca96abac8a
Set up basic test gh worklows ( #79 )
...
* Set up basic test gh worklows
* update file paths
* env check
* add env check
* Update requirements.txt
* simplify
* upd
2024-03-18 11:58:37 -05:00
rasbt
4fc6de7afa
add notes
2024-03-17 09:29:06 -05:00
rasbt
d60da19fd0
add more notes and embed figures externally to save space
2024-03-17 09:08:38 -05:00
Intelligence-Manifesto
d4b4e3d0f0
the above -> the following
2024-03-15 05:00:28 +08:00
rasbt
1870b4bacd
update stride param
2024-03-13 08:39:59 -05:00
rasbt
244137e8a1
amend
2024-03-10 08:05:22 -05:00
rasbt
76205521d7
different dropout behavior on macos and linux
2024-03-10 07:58:10 -05:00
rasbt
73822b8bfa
move ex 3.3 solution outside main chapter
2024-03-10 07:18:24 -05:00
rasbt
da33ce8054
remove redundant unsqueeze in mask
2024-03-09 17:42:31 -06:00
rasbt
3beaea46ce
add lowres figs for better navigation
2024-03-08 07:18:06 -06:00
rasbt
b6fe1a37b3
also add simple wrapper
2024-03-06 08:38:53 -06:00
rasbt
87fcfd9245
mha variants
2024-03-06 08:30:32 -06:00
rasbt
d4754f1bdd
change dim=1 to dim=-1
2024-03-04 18:54:43 -06:00
rasbt
b827bf4eea
remove redundant double-unsequeeze
2024-02-29 08:31:07 -06:00
rasbt
250e6306e2
use attn_scores from sec 3.4 instead of 3.3
2024-02-14 20:23:59 -06:00
Intelligence-Manifesto
6a09e7b03a
**step 2**
...
step 2: According to the context, the formatting here should be **step 2**.
Additionally, it seems that there is a lack of text description for step 1 in this section, as other sections are all labeled with steps 1, 2, 3 in order, clearly indicating the steps.
2024-02-12 18:32:28 +08:00
Intelligence-Manifesto
1278615c25
12 -> 21
...
12 -> 21
2024-02-11 14:17:55 +08:00
rasbt
3a5fc79b38
add and update readme files
2024-02-05 06:51:58 -06:00
rasbt
8860e16e05
<|endoftext|> token in dataset v1
2024-01-21 12:03:04 -06:00
rasbt
92896d817c
add toggle for qkv_bias
2024-01-17 07:50:57 -06:00
rasbt
dfe2c3b46f
use blocksize in positional embedding
2024-01-15 08:15:33 -06:00
rasbt
9e85f13ba9
readability improvements
2024-01-15 07:36:19 -06:00
