Commit History
fix: torch_dtype mistral default to fp32 (#1050)
c3e8165
unverified
Update FUNDING.yml for Kofi link (#1067)
7f38175
unverified
Sponsors (#1065)
1496441
unverified
Phi2 rewrite (#1058)
732851f
unverified
streaming multipack for pretraining dataset (#959)
553c80f
unverified
Update tests-docker.yml (#1052) [skip ci]
eb4c994
unverified
feat: always push checkpoint to hub if set (#1049) [skip ci]
cbdbf9e
unverified
feature: better device mapping for large models (#918)
bdfefaf
unverified
set default for merge (#1044)
63fb3eb
unverified
fix model card upload for PEFT models (#1043)
31d2350
unverified
RL/DPO (#935)
f243c21
Added chatglm3 conversation type for training models like TinyLLama (#1036)
59b2d30
unverified
bump transformers and update attention class map name (#1023)
bcc78d8
unverified
chore(config): clean up old log for Qwen (#1034)
74532dd
unverified
fix: lint (#1037)
8ba27f3
unverified
chore(readme): update instruction to set config to load from cache (#1030)
b31038a
unverified
added tiny llama examples for lora and qlora (#1027)
c75f916
unverified
Tim Dolan
commited on
use recommended setting for use_reentrant w gradient checkpointing (#1021)
4d2e842
unverified
Fix: bf16 support for inference (#981)
3678a6c
unverified
Adds chat templates (#1022)
f8ae59b
unverified
[WandB] Push axolotl config to top level wandb files (#1014)
4f4d638
unverified
add ultrachat prompt strategies (#996)
ba043a3
unverified
feat: remove need to add load_in* during merge (#1017)
f6ecf14
unverified
[Docs] Nit: Remind people to auth to wandb if they are going to use it (#1013)
dec66d7
unverified
Update README.md (#1012)
76357dc
unverified
remove landmark attn and xpos rope implementations (#1010)
70b46ca
unverified
add config to model card (#1005)
85dd4d5
unverified
Set eval_sample_packing to false in mistral config.yaml (#1003)
384b817
unverified
Kevin Sydney
commited on
FEAT: add tagging support to axolotl (#1004)
db9094d
unverified
Add an example config for finetuning a 34B model on a 24GB GPU (#1000)
6ef46f8
unverified
Evan Griffiths
commited on
set output_router_logits for mixtral config: (#995)
628b754
unverified
support for cuda 12.1 (#989)
37820f6
unverified
chore: Update transformers to latest (#986)
7d4185f
unverified
change val size (#992)
93ebec1
unverified
Add tests to Docker (#993)
2e61dc3
unverified
Feat: Warns to add to modules_to_save when adding tokens or switching special_tokens (#787)
1ffa386
unverified
bump actions versions
62ba160
fix mistral prompt assembly (#982)
7bbaac9
unverified
Dockerfile torch fix (#987)
161bcb6
unverified
Update README.md (#966)
d25c34c
unverified
fix: add lr scheduler kwargs to Trainer (#972)
13e9381
unverified
fix for build for nccl in dockerfile (#970)
85de004
unverified
update to latest nccl in docker image (#965)
80ec7af
unverified
update transformers to fix checkpoint saving (#963)
f28e755
unverified
dumpmemory
commited on