 k.h.lai
|
30e70334f7
llava-cli: fix base64 prompt (#7248)
|
il y a 1 an |
 Johannes Gäßler
|
1c570d8bee
perplexity: add BF16 vs. FP16 results (#7150)
|
il y a 1 an |
 Neo Zhang
|
948f4ec7c5
[SYCL] rm wait() (#7233)
|
il y a 1 an |
 Joan Fontanals
|
9aa672490c
llama : rename jina tokenizers to v2 (#7249)
|
il y a 1 an |
 Brian
|
b1f8af1886
convert.py: Outfile default name change and additional metadata support (#4858)
|
il y a 1 an |
 Benjamin Findley
|
e586ee4259
change default temperature of OAI compat API from 0 to 1 (#7226)
|
il y a 1 an |
|
Neo Zhang
|
cbf75894d2
[SYCL] Add oneapi runtime dll files to win release package (#7241)
|
il y a 1 an |
|
Neo Zhang
|
0d5cef78ae
[SYCL] update CI with oneapi 2024.1 (#7235)
|
il y a 1 an |
|
Johannes Gäßler
|
dc685be466
CUDA: add FP32 FlashAttention vector kernel (#7188)
|
il y a 1 an |
 Georgi Gerganov
|
6f1b63606f
cmake : fix version cmp (#7227)
|
il y a 1 an |
 slaren
|
b228aba91a
remove convert-lora-to-ggml.py (#7204)
|
il y a 1 an |
|
Georgi Gerganov
|
7bd4ffb780
metal : fix warnings (skipme) (#0)
|
il y a 1 an |
|
Georgi Gerganov
|
1622ac023f
sync : ggml
|
il y a 1 an |
|
Georgi Gerganov
|
6aeff24f8b
metal : fix indent (ggml/0)
|
il y a 1 an |
|
Georgi Gerganov
|
325756d28d
ggml : resolve merge (ggml/0)
|
il y a 1 an |
 Josh Ramer
|
fed0108491
Scripting & documenting debugging one test without anything else in the loop. (#7096)
|
il y a 1 an |
 Xuan Son Nguyen
|
72c177c1f6
fix system prompt handling (#7153)
|
il y a 1 an |
 compilade
|
5a419926b0
convert-hf : support bfloat16 conversion (#7158)
|
il y a 1 an |
|
Georgi Gerganov
|
fae9d234b6
sync : ggml
|
il y a 1 an |
 Justina Cho
|
f5ef34e428
feat: implemented sigmoid function (ggml/806)
|
il y a 1 an |
 Borislav Stanimirov
|
ef0d5e3ec9
build: fix and ignore msvc warnings (ggml/805)
|
il y a 1 an |
 CrispStrobe
|
3292733f95
convert : skip unaccessible HF repos (#7210)
|
il y a 1 an |
 Steve Grubb
|
988631335a
server : free llama_batch on exit (#7212)
|
il y a 1 an |
 Haoxiang Fei
|
f99e1e456e
llama : lookup word in vocab before doing BPE merges (#7193)
|
il y a 1 an |
|
Johannes Gäßler
|
5ae3426b0b
server: fix reported top tokens for temperature 0 (#7203)
|
il y a 1 an |
 Joan Fontanals
|
b83cc3f5b3
llama : add Jina Embeddings architecture (#6826)
|
il y a 1 an |
|
Georgi Gerganov
|
9cb317f77e
ggml : full ALiBi support (#7192)
|
il y a 1 an |
|
slaren
|
e849648888
llama-bench : add pp+tg test type (#7199)
|
il y a 1 an |
|
Georgi Gerganov
|
18e437665c
metal : fix flash attention kernel requirements (#7169)
|
il y a 1 an |
|
Georgi Gerganov
|
8c660242d7
convert : print "ignore_merges" field
|
il y a 1 an |
