Pyry Takala
28fd25b2bc
Merge 26b0941f75 into 8630d32be4
2026-01-26 21:56:33 -08:00
Sofie Van Landeghem
d4ea28d4e2
Fix args in readme ( #438 )
...
* fix commands in readme, using new arg format
* fix typo
* add required -i flag to chat_eval example runs
2026-01-15 16:26:38 -08:00
Andrej
d5759400f9
fixing two typos in comments
2025-12-08 20:03:08 -08:00
Sofie Van Landeghem
26b0941f75
fix
2025-11-22 22:51:42 +01:00
Sofie Van Landeghem
c09b897601
further cleanup
2025-11-22 22:50:34 +01:00
Sofie Van Landeghem
df9a644e24
make code bit more succinct
2025-11-22 22:48:55 +01:00
Pyry Takala
a33d04dca1
Cap stop parameter and warn once when it exceeds dataset size
2025-11-21 20:51:46 +00:00
Pyry Takala
85e49943ed
Gracefully handle stop > dataset_size with warning
2025-11-21 20:04:33 +00:00
Pyry Takala
cd782a1977
Fix: Validate stop parameter against dataset size
...
Add validation in Task.__len__() to ensure stop parameter does not exceed
the actual dataset size. This prevents IndexError crashes during training
when invalid stop values are provided.
The validation is centralized in the base Task class and preserves the
original lazy evaluation behavior - num_examples() is only called when
needed (for validation when stop is provided, or for default value when
stop is None).
Fixes issue where training would crash with IndexError when iterating
over Task instances with stop > dataset_size.
2025-11-20 04:18:42 +00:00
Eric Silberstein
024781f9df
fixing two typos in comments
2025-11-19 15:12:53 -05:00
Eric Silberstein
97770700f2
change test/train split approach because random.seed(1) and random.seed(-1) do the same thing
2025-11-19 14:51:02 -05:00
svlandeg
c72b8b2309
add explicit UTF-8 encoding
2025-11-03 21:27:12 +01:00
Andrej Karpathy
05a051dbe9
fix tokenization bug, there should be no space before first letter. sigh
2025-10-24 15:06:06 +00:00
Andrej Karpathy
8892470f29
add the SpellingBee task so that nanochat can count r in strawberry etc. along the way we had to add a bunch of new functionality, e.g. extend the calculator to support the count function of python. possibly the current TaskMixture uses way too many synthetic examples of SpellingBee because the eval gives us exactly 100% performance on spelling. We can tune this later to reclaim some wall clock time here I think
2025-10-24 14:02:48 +00:00
Andrej Karpathy
03cddd9878
actually let's not brick code on git pull. change error to warning
2025-10-21 15:13:25 +00:00
Andrej Karpathy
fe5aed940b
add personality to nanochat. breaks previous code on git pull and requires download of a new file from s3, but there is a helpful error message so hopefully its ok
2025-10-21 15:04:58 +00:00
karpathy
3a5e0bc50b
initial commit
2025-10-13 06:49:24 -07:00
