 AlpinDale
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
hai 11 meses |
 50h100a
|
d5dbd29db4
hoist sampler internals into a single function.
|
hai 11 meses |
|
AlpinDale
|
da223153c6
feat&fix: cohere support and missing GPU blocks (#333)
|
hai 1 ano |
|
AlpinDale
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
hai 1 ano |
|
AlpinDale
|
9fa99215f8
feat: add cubic sampling (#280)
|
hai 1 ano |
|
AlpinDale
|
657aec0cbd
refactor: OpenAI endpoint (#261)
|
hai 1 ano |
|
AlpinDale
|
4d04ade9ef
feat: fine-grained seeds (#279)
|
hai 1 ano |
 anon998
|
35b9033782
fix: crash in quadratic sampling when batch > 1 (#253)
|
hai 1 ano |
 50h100a
|
f619c96c79
fix: zero token output due to temperature bias (#243)
|
hai 1 ano |
|
50h100a
|
53a9c60442
fix: logit processor declarations and application (#242)
|
hai 1 ano |
|
AlpinDale
|
e73a92ad2f
fix: remove the mask for quadratic sampling (#236)
|
hai 1 ano |
|
AlpinDale
|
1c46fa31ad
feat: add quadratic sampling (#233)
|
hai 1 ano |
|
AlpinDale
|
c3a221eb02
feat: GGUF, QuIP#, and Marlin support (#228)
|
hai 1 ano |
|
AlpinDale
|
c0aac15421
feat: S-LoRA support (#222)
|
hai 1 ano |
|
AlpinDale
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
hai 1 ano |
 Stefan Gligorijevic
|
9e7e108dc8
chore: clamp dynatemp_min (#214)
|
hai 1 ano |
|
Stefan Gligorijevic
|
56446a04bb
feat: dynamic temperature (#209)
|
hai 1 ano |
|
AlpinDale
|
d54791aaa8
feat: reduce sampler overhead by making it less blocking (#198)
|
hai 1 ano |
|
AlpinDale
|
653da510d1
chore: rewrite InputMetadata (#143)
|
hai 1 ano |
|
AlpinDale
|
6c914ea0e4
fix: `SequenceOutputs` -> `SequenceOutput` (#133)
|
hai 1 ano |
|
Stefan Gligorijevic
|
a5255901c6
feat: min_p sampling (#106)
|
hai 1 ano |
 AlpinDale
|
ae7d8df224
fix lint issues (again)
|
hai 1 ano |
|
50h100a
|
fa0ae5a2c9
feat: new mirostatv2 implementation (#96)
|
hai 1 ano |
|
AlpinDale
|
69204736de
Revert "fix: sync CPU delay in sampler (#93)"
|
hai 1 ano |
|
AlpinDale
|
ce66e1df56
fix: sync CPU delay in sampler (#93)
|
hai 1 ano |
|
AlpinDale
|
74604eb252
fix: pylint complaints (#91)
|
hai 1 ano |
|
AlpinDale
|
efc6f7fbec
chore: reformats (#90)
|
hai 1 ano |
|
AlpinDale
|
3d72f05c7b
feat: flattened 1D tensor -> 2D tensor (#85)
|
hai 1 ano |
|
AlpinDale
|
e6be0118c9
feat: prompt logprobs and batched samplers (#77)
|
hai 1 ano |
|
Stefan Gligorijevic
|
34c1c8c83a
feat: Enable banning tokens (#80)
|
hai 1 ano |
