qazalandGitHub 73b067f5ce Bitcast p2 bfloat16 tests + clang fix (#2635) ...

* add bf16 test support

this model takes me almost a minute to download though:

https://huggingface.co/TinyPixel/Llama-2-7B-bf16-sharded/resolve/main/pytorch_model-00001-of-00014.bin?download=true: 100%|█████████████████████████████| 981M/981M [00:40<00:00, 24.2MB/s]

* ensure we first load if it is bitcast to avoid taking the address of an rvalue

* tiny bf16 in the cloud

skip GPU

* should skip torch

lint

* Revert "ensure we first load if it is bitcast to avoid taking the address of an rvalue"

This reverts commit b86a28ab84.

* break the kernel

* skip LLVM and GPU in CI

* skip CUDA

2023-12-08 10:30:10 -08:00

..

codegen

FROM -> COPY, move vars_from_ast (#2675)

2023-12-07 16:32:30 -08:00

features

FROM -> COPY, move vars_from_ast (#2675)

2023-12-07 16:32:30 -08:00

nn

Bitcast p2 bfloat16 tests + clang fix (#2635)

2023-12-08 10:30:10 -08:00

renderer

green more dtypes tests (#2656)

2023-12-08 10:29:20 -08:00

runtime

multitensor start (#2676)

2023-12-07 17:07:05 -08:00

shape

use one kvcache tensor in gpt2 instead of two separate caches (#2662)

2023-12-06 20:59:17 -05:00

__init__.py

move device to device.py (#2466)

2023-11-27 11:34:37 -08:00

device.py

multitensor start (#2676)

2023-12-07 17:07:05 -08:00

graph.py

multitensor start (#2676)

2023-12-07 17:07:05 -08:00

helpers.py

GPT bugfixes (#2624)

2023-12-05 11:42:28 -08:00

jit.py

split large jit into several graphs (#2650)

2023-12-07 10:58:25 -08:00

lazy.py

FROM -> COPY, move vars_from_ast (#2675)

2023-12-07 16:32:30 -08:00

mlops.py

binaryops xor (#2627)

2023-12-05 13:21:42 -08:00

ops.py

FROM -> COPY, move vars_from_ast (#2675)

2023-12-07 16:32:30 -08:00

realize.py

FROM -> COPY, move vars_from_ast (#2675)

2023-12-07 16:32:30 -08:00

tensor.py

masked shapetrackers (#2657)

2023-12-06 11:22:26 -08:00