Morph MORPH
SIGN IN SIGN UP
dhruvhead / open-instruct UNCLAIMED
0 0 0 Python

COMMITS

/ scripts/submit_eval_jobs.py
docs
July 5, 2024
H
remove jupiter from default
Hamish Ivison committed
d5cd80c
June 27, 2024
H
Add nfs guard
Hamish Ivison committed
9f9c62a
H
remove nfs if jupiter or asked
Hamish Ivison committed
e6e8531
June 24, 2024
H
no quant anymore.
Hamish Ivison committed
f374e55
H
add comment on eot
Hamish Ivison committed
87f28f1
H
additional stop option for llama 3 instruct
Hamish Ivison committed
9913fc1
June 19, 2024
H
Merge branch 'main' into llama-3-evals
Hamish Ivison committed
ca5dcd0
May 29, 2024
N
basic structure of preemptible jobs on beaker
Nathan Lambert committed
b6a1322
April 5, 2024
H
bump alpaca eval
Hamish Ivison committed
6f43970
D
Add a flag for olmo models.
dwadden committed
dbb51fc
D
Replace tokenizer path.
dwadden committed
b06f087
May 18, 2024
M
update submit_eval_jobs.py
mickelliu committed
8925430
May 9, 2024
D
Fix typo.
dwadden committed
71e3a57
May 8, 2024
D
Refactor GPU and batch size adjustment code.
dwadden committed
5ed304d
April 30, 2024
D
Add trailing backslash in GSM eval command.
dwadden committed
aabd6e8
April 29, 2024
D
Add GSM double-newline flag to submission script.
dwadden committed
58f92c8
April 26, 2024
N
up
Nathan Lambert committed
eaeef71
April 25, 2024
N
up
Nathan Lambert committed
aa12177
April 12, 2024
O
Add --experiments argument
Oyvind Tafjord committed
a9c5414
April 6, 2024
D
Set max new tokens=2048 for OLMo alpaca_eval.
dwadden committed
f57015e
March 29, 2024
Y
Add IFEval for formatting instruction following and Xstest evaluation for refusal (#133)
Yizhong Wang committed
ecf0c50
March 5, 2024
D
Add back comment describing `model_info` tuple.
dwadden committed
581d0ee
February 27, 2024
D
Run all tasks in a single experiment.
dwadden committed
be4d9df
D
Add option to pass different beaker image.
dwadden committed
e50f9b4
D
Enable evaluation of checkpoints in subfolders.
dwadden committed
5cbacb6
February 21, 2024
H
fix2
Hamish Ivison committed
1081dba
H
fix
Hamish Ivison committed
89583d3
H
update eval script with clusters etc
Hamish Ivison committed
2249e5d
H
tokenizer util
Hamish Ivison committed
7f7acd4
February 13, 2024
Y
Update TruthfulQA evaluation using retrained LLaMa judge models. (#112)
Yizhong Wang committed
2b6780c
H
Beaker improvements & HF formatting (#110)
Hamish Ivison committed
8dadfdf
January 29, 2024
H
fix name of olmo chat template
Hamish Ivison committed
09d567c
H
remove my silly eval job
Hamish Ivison committed
0a1c13b
H
optional olmo import
Hamish Ivison committed
a091092
December 19, 2023
H
Fixes and changes for olmo
Hamish Ivison committed
2771cb2
November 17, 2023
Y
Fix several issues in the evals. (#79)
Yizhong Wang committed
4a2e9dd
October 29, 2023
Y
Evals for more models.
yizhongw committed
a8780d7
October 16, 2023
Y
Update ids for model checkpoints.
yizhongw committed
578b1a5
October 10, 2023
Y
update eval job submission script.
yizhongw committed
3faa603
September 4, 2023
Y
Use import_module to dynamically import the chat formatting function, allowing more flexibility for user-defined formatting.
yizhongw committed
1a35973
September 3, 2023
Y
Remove mgsm and superni from the evaluation set, since they are not used.
yizhongw committed
2c54b5e
Y
Add toxigen eval to automatic eval submisstion.
yizhongw committed
53f8a65
Y
Use `chat_formatting_function` to specify which prompt encoding function the evals should use.
yizhongw committed
5d6264a
September 2, 2023
Y
Support LLaMa2 chat format.
yizhongw committed
ed2cded
June 9, 2023
H
initial commit
Hamish Ivison committed
e44fcf5