| .. |
|
gguf
|
fcfb72af24
Support arbitrary model in GGUF. (#381)
|
9 mēneši atpakaļ |
|
__init__.py
|
07aa2a492f
upstream: add option to specify tokenizer
|
1 gadu atpakaļ |
|
block.py
|
ac82b67f75
feat: naive context shift and various QoL changes (#289)
|
10 mēneši atpakaļ |
|
config.py
|
bd0ddf1cfe
feat: EETQ quantization (#408)
|
9 mēneši atpakaļ |
|
grammar.py
|
0527131e93
fix: grammar logits processor (#268)
|
10 mēneši atpakaļ |
|
logger.py
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
9 mēneši atpakaļ |
|
logits_processor.py
|
3df36ee07d
fix: logit bias logitproc (#278)
|
10 mēneši atpakaļ |
|
outputs.py
|
c18bf116da
fix stop strings not being excluded from outputs
|
9 mēneši atpakaļ |
|
sampling_params.py
|
c18bf116da
fix stop strings not being excluded from outputs
|
9 mēneši atpakaļ |
|
sequence.py
|
c18bf116da
fix stop strings not being excluded from outputs
|
9 mēneši atpakaļ |
|
test_utils.py
|
50c2434267
move megatron to a top-level directory
|
9 mēneši atpakaļ |
|
utils.py
|
531969a0b2
move merge_async_iterators to common utils
|
9 mēneši atpakaļ |
sgsdxzy
fcfb72af24
Support arbitrary model in GGUF. (#381)