DeepSpeed
Fixing several bugs in the inference-api and the kernels
#1951
Merged

Fixing several bugs in the inference-api and the kernels #1951

jeffra merged 15 commits into master from fix-gpt-inference
RezaYazdaniAminabadi
fixing several bugs in the inference-api and the kernels
7707cbce
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from jeffra jeffra 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from samyam samyam 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from tjruwase tjruwase 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from ShadenSmith ShadenSmith 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from conglongli conglongli 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from awan-10 awan-10 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from cli99 cli99 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from eltonzheng eltonzheng 3 years ago
RezaYazdaniAminabadi RezaYazdaniAminabadi requested a review from minjiaz minjiaz 3 years ago
jeffra formatting
1aa5816d
jeffra add hf inference unit test
7447cee3
jeffra update unit tests for fp32/16
a2edb1d3
jeffra skip test for now, needs debugging
773a2581
jeffra Merge branch 'master' into fix-gpt-inference
5fcfe1b0
jeffra Merge branch 'master' into fix-gpt-inference
fc868e3a
fix rotary_pos embedding for GPT-J inderence
6908fb15
Merge branch 'fix-gpt-inference' of github.com:microsoft/DeepSpeed in…
f931bb91
jeffra turn unit test back on
94a6ca43
jeffra fix for gpt-j fp16 ckpt revision
af24e619
jeffra add injection workaround for pt1.2
29c2a246
jeffra fix for amd
d290b019
jeffra fix for megatron + pt1.2
5dce777b
jeffra skip on torch 1.2, several issues there
51e5c012
jeffra
jeffra approved these changes on 2022-05-24
jeffra jeffra merged 8164ea9e into master 3 years ago
mrwyattii mrwyattii deleted the fix-gpt-inference branch 2 years ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone