d314e96aa2
formatting
2025-12-09 12:48:46 +01:00
bbc57da7d5
slightly nicer error message
2025-12-09 12:46:48 +01:00
f1bf69d562
feat: pad vocab size to 64 for DDP optimizers and efficiency
2025-12-09 12:38:18 +01:00
7931e0903a
rename checkpoint_dir to checkpoints_dir for consistency.
2025-12-08 18:32:12 -08:00
849d95ae1f
remove unnecessary check to make the logic in CausalSelfAttention.forward() clearer
2025-12-08 18:30:37 -08:00
1b2a675c88
Improve KV cache code readability
2025-12-08 18:19:05 -08:00
72a7cf2bc4
Fix distributed Parquet dataloader resume for multi-epoch training
2025-12-08 18:15:02 -08:00
bffdb2ef91
group common code to make things neater in gpt logit computation
2025-12-09 02:01:05 +00:00
cbf30c842c
apply float32 cast before logits softcapping so the tanh is in fp32. torch compile fuses this correctly with no extra memory costs.
2025-12-08 14:17:43 -08:00
90442de35f
fix bug where any rank has to be able to create checkpoint_dir if saving optim
2025-12-08 20:45:19 +00:00
01ea71be39
Fix distributed Parquet dataloader resume for multi-epoch training
2025-12-08 00:10:19 -06:00
06677c30e0
Refactor dimension validation for KV cache
2025-11-28 15:22:18 -05:00
a770dcef2e
Fix kv_cache indexing to explicitly include head dimension
2025-11-28 15:00:14 -05:00
16788eed3c
fix(model): apply float32 cast before logits softcapping
...
This change ensures that the logits softcapping operation (tanh) is performed in float32 precision rather than bfloat16. Previously, the code cast to float32 after the tanh operation, which meant the non-linearity was computed with bfloat16 precision
2025-11-23 20:12:09 +05:30
5c93a56be5
remove unnecessary check
2025-11-19 16:31:41 -05:00
a4a0959c73
renamed find_largest_model() argument checkpoint_dir to checkpoints_dir for clarity
2025-11-19 15:33:36 -05:00
11e68bf442
Fix comment: rotary embeddings final dimension size
2025-11-17 11:32:56 -05:00
bc1fca39f3
mqa -> gqa to reduce confusion
2025-11-15 15:43:37 +00:00
f66a780f68
Fix torch.dtype mismatching when running engine inline test.
2025-11-14 07:28:29 -08:00
4763ce612a
Small fixes to typos
2025-11-14 07:25:59 -08:00
c6f5bd67db
revert change of base to sft for quick inline test
2025-11-14 12:20:03 +01:00
a2fb3c83a6
fix typos
2025-11-14 11:20:25 +01:00
c6abcdfe3a
big change: add pretraining resumption logic so that checkpoints can now be approximately resumed and training can continue. this is useful for very long runs when you don't want the anxiety of your run crashing for some reason. alternatively, it's a way to recover training in the event of loss spikes. i mean, this should have been there in v0 but it's ok. the resumption is approximate to control complexity and bloat, but it's possible we want to change that in the future. to use, set --save_every to a step interval to write checkpoints with, and then use --resume_from_step to resume optimization from a given step. only base model training (pretraining) supports this atm, but it's ok because midtraining is comparably quite a bit faster.
2025-11-13 15:34:40 +00:00
91f09ccd0d
minor fix comment in engine
2025-11-13 15:28:18 +00:00
b399e43168
fix engine test bug
2025-11-06 08:56:45 +08:00
3a2ae631c4
Merge branch 'master' into master
2025-11-04 16:35:02 -08:00
d1558c7873
handle bf16 on MPS by casting to fp32 during load checkpoint
2025-11-04 09:42:50 -08:00
1e89af9862
Replace fcntl with filelock for Windows compatibility
2025-11-04 07:22:34 +00:00
7a40ee77b4
fix: cast bf16 to fp32 on MPS (like CPU) to avoid dtype issues
2025-11-03 16:00:56 -05:00
2ce62ec076
ensure consistency of quotes within each statement
2025-11-03 21:52:02 +01:00
c72b8b2309
add explicit UTF-8 encoding
2025-11-03 21:27:12 +01:00
f1e15f5f4d
Fixing subtle bug: lstrip removes all matching characters, including potentially required ones. Use removeprefix instead.
2025-11-02 23:40:37 -06:00
b6da6982f6
fix nanochat logo: the t was placed too far to the right
2025-11-02 08:17:00 -08:00
d1ac0b2d07
when loading models on CPU, convert tensors from bfloat16 to float
2025-11-02 07:58:56 -08:00
5bfcd31b73
revert more formatting changes
2025-11-02 14:17:10 +01:00
036a3c5881
revert formatting changes to facilitate review
2025-11-02 14:16:43 +01:00
d54c9cbf8c
CPU Support, as bfloat16 params breaks inference
2025-11-01 23:38:50 +01:00
cf587acb1a
move eval bundle download to be lazy and inside the python code so that we can substantially simplify the run bash scripts
2025-11-01 16:04:38 +00:00
ad39db5a23
tiny fix to comment
...
Update engine.py with correct error message on assert
2025-11-01 07:43:57 -07:00
630f54ae5a
use empty locals and globals in call to eval() in engine tool use
...
harden eval: prevent the calc tool from accessing globals and locals
2025-11-01 07:22:59 -07:00
1ccbaf4416
nit delete redundant catch/raise in execute
...
Remove redundant exception handling in chdir
2025-10-29 08:10:03 -07:00
29ff38d94b
Merge pull request #35 from bhaskar0210s/master
...
fix: return inf instead of crashing when evaluate_bpb has zero total_bytes
2025-10-29 08:06:24 -07:00
b996131570
Merge branch 'master' into logo/kerning-update
2025-10-29 11:45:40 +01:00
3fa974f93c
few more reverts
2025-10-29 11:45:02 +01:00
cbd560a83d
revert formatting changes to minimize diff and merge conflicts
2025-10-29 11:42:56 +01:00
2b9c085559
update the kv_shape
2025-10-27 02:47:13 -07:00
b062b422ac
Fix kv cache, given resize will destroys the logical structure
2025-10-27 02:23:08 -07:00
fca2b8cd07
harden eval: prevent the calc tool from accessing globals and locals
...
By passing empty globals() and locals() to eval() we can prevent simple
malicious cases where the user gets the model to output something like
```<global variable/func> or "a".count("a")```
e.g.
```signal.raise_signal(9) or "a".count("a")``` which would kill the process.
or one could maybe get it to output secrets etc.
I think to make it 100% secure one would need to parse the AST and only execute secure nodes but this should make it much more robust.
2025-10-24 14:41:12 -05:00
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
cc3636b01c
allow the tokenizer visualize_tokenization to also print the exact token id. you can never be paranoid enough
2025-10-24 13:27:05 +00:00
Matěj Kripner