• Joined on 2024-05-31
tacit synced commits to refs/pull/13/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 25 commits »
tacit synced and deleted reference refs/tags/refs/pull/61/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
tacit synced and deleted reference refs/tags/refs/pull/50/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
tacit synced and deleted reference refs/tags/refs/pull/133/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
tacit synced and deleted reference refs/tags/refs/pull/132/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
tacit synced and deleted reference refs/tags/refs/pull/126/merge at tacit/nanochat from mirror 2025-10-22 14:22:13 +00:00
tacit synced commits to refs/pull/94/merge at tacit/nanochat from mirror 2025-10-22 06:12:15 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
49cd02f283 fix: remove unnecessary tensor allocation in DistAdamW optimizer
Compare 3 commits »
tacit synced commits to refs/pull/91/head at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
067298c51b reverted back
tacit synced commits to refs/pull/90/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/64/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/46/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/18/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/142/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
7a52f9bfbb Updates lockfile with CPU package support without overwriting other architectures
760af62e11 Git ignore eval_bundle
901b075605 Fix GPU-less CPU use on Linux with specific Torch indexes
Compare 4 commits »
tacit synced commits to refs/pull/91/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
067298c51b reverted back
Compare 2 commits »
tacit synced commits to refs/pull/142/head at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
7a52f9bfbb Updates lockfile with CPU package support without overwriting other architectures
760af62e11 Git ignore eval_bundle
901b075605 Fix GPU-less CPU use on Linux with specific Torch indexes
Compare 3 commits »
tacit synced commits to refs/pull/106/merge at tacit/nanochat from mirror 2025-10-22 06:12:14 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
49cd02f283 fix: remove unnecessary tensor allocation in DistAdamW optimizer
Compare 3 commits »
tacit synced commits to refs/pull/98/merge at tacit/nanochat from mirror 2025-10-21 22:02:20 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/97/merge at tacit/nanochat from mirror 2025-10-21 22:02:20 +00:00
2e938530ce delete spurious torch.empty allocation in adamw
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
Compare 28 commits »
tacit synced commits to refs/pull/94/merge at tacit/nanochat from mirror 2025-10-21 22:02:20 +00:00
a088b7a6ec use enable_gqa of pytorch sdpa, allows us to delete some code, didnt realize it's available
94ee507054 quick fix base eval due to fewshot requirement
33e8a27f91 Merge karpathy/cpu-mps-dev , adding the ability to run on CPU, on MPS, or on CUDA, with autodetect. Gnarly PR, nonzero chance I broke something.
50bea28ef9 also add readme mention of the cpu mps changes
Compare 26 commits »
tacit synced commits to refs/pull/86/merge at tacit/nanochat from mirror 2025-10-21 22:02:19 +00:00
03cddd9878 actually let's not brick code on git pull. change error to warning
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
Compare 3 commits »