Commit History

always drop samples that are too long (#452)
50682a3
unverified

winglian commited on

set env var for FSDP layer to wrap (#453)
5a1985b
unverified

winglian commited on

Merge pull request #451 from OpenAccess-AI-Collective/eval-is-causal
5e9c6af
unverified

tmm1 commited on

is_causal fix for evals?
fbf49a4

winglian commited on

add missing positional arg (#450)
58cf7e7
unverified

winglian commited on

feat(docs): improve user customized prompts (#443)
04a42b6
unverified

Nanobit commited on

feat(doc): add pillow to lambda instructions (#445)
919f4ca
unverified

Nanobit commited on

fix evals (#447)
ee26281
unverified

winglian commited on

gracefully handle empty input (#442)
9d629d8
unverified

winglian commited on

support user defined prompters, pretokenized datasets in config, local parquet, local arrow files (#348)
d2e7f27
unverified

winglian commited on

docs(readme): add `cd axolotl` (#440)
d21318d
unverified

philpax commited on

disable eval using multipack for now (#437)
f733d0f
unverified

winglian commited on

fix comma, not a tuple (#436)
008505c
unverified

winglian commited on

use save_strategy from config if available (#434)
b3f5e00
unverified

winglian commited on

set env for FSDP offload params (#433)
5247c50
unverified

winglian commited on

flash attn pip install (#426)
cf66547
unverified

mhenrichsen Ubuntu mhenrichsen Mads Henrichsen winglian commited on

standardize attn hijack patches (#381)
06edf17
unverified

tmm1 winglian commited on

remove extra accelearate in requirements (#430)
82e111a
unverified

winglian commited on

fix fixture for new tokenizer handling in transformers (#428)
8cace80
unverified

winglian commited on

fix orca prompts (#422)
1b7e860
unverified

winglian commited on

Fix(docs): Remove gptq+lora and fix xformer compat list (#423)
3d1f203
unverified

Nanobit commited on

just resort to tags ans use main-latest (#424)
d3d6fd6
unverified

winglian commited on

Fix(template): Inform to place stack trace to Issue (#417)
b7449a9
unverified

Nanobit winglian commited on

use inputs for image rather than outputs for docker metadata (#420)
5f80b35
unverified

winglian commited on

hopefully improve the README (#419)
2495909
unverified

winglian commited on

tag with latest as well for axolotl-runpod (#418)
7af8166
unverified

winglian commited on

Merge pull request #413 from mhenrichsen/chore/update-deepseed-config
f806e86
unverified

mhenrichsen commited on

Feat(doc): Add lr_quadratic_warmup to readme (#412)
2b990eb
unverified

Nanobit commited on

update path to align with fsdp example
bd8cab4

mhenrichsen commited on

Fix(config): Update handling of deepspeed config (#404)
c01015f
unverified

Nanobit commited on

Fix(docs): Update flash attn requirements (#409)
72fe3f8
unverified

Nanobit commited on

update docs for tokenizer_legacy (#401)
47961fd
unverified

winglian commited on

Fix(template): Remove iPhone/android from Issue template (#407)
7ad37cb
unverified

Nanobit commited on

Ax art (#405)
29241cf
unverified

winglian commited on

add templates, CoC and contributing guide (#126)
31db0ec
unverified

lightningRalf winglian Nanobit commited on

fix eval steps and strategy (#403)
da10af0
unverified

winglian commited on

better handling of empty input ids when tokenizing (#395)
85cf4f8
unverified

winglian commited on

add utils.data.prepare_dataset
2e22404

tmm1 commited on

Feat(doc): Add how to save by epochs (#396)
be294fd
unverified

Nanobit commited on

use context manager to run things on rank0 before others (#397)
fc2d6be
unverified

winglian commited on

don't use mask expansion for inference (#392)
1687be6
unverified

winglian commited on

Feat(doc): Add max_steps to readme (#389)
41ecb45
unverified

Nanobit commited on

Feat(config): add max steps (#387)
3c2ad00
unverified

ittailup commited on

Added "epoch" evaluation_strategy (#388)
5d48a10
unverified

flotos commited on

Feat(config): Add hub_strategy (#386)
73a0b6e
unverified

Nanobit commited on

Error msg for sharegpt if conv has less than 2 msg (#379)
63fdb5a
unverified

flotos commited on

new llama-2 default settings (#370)
fdffef5
unverified

mhenrichsen Mads Henrichsen commited on

don't pass rope_scaling kwarg if it's None (#383)
919246f
unverified

winglian commited on