Giters
facebookresearch
/
metaseq
Repo for external large-scale work
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
6411
Watchers:
108
Issues:
292
Forks:
719
facebookresearch/metaseq Issues
Weights/Code for CM3Leon
Updated
3 months ago
Comments count
2
How to load the checkpoints into a HF model?
Updated
4 months ago
How to generate binary data file ?
Closed
a year ago
Comments count
1
I change Num_head of OPT-1.3b,and it cause CUDA Error: IndexSelectLargeIndex,
Updated
10 months ago
Access request for opt-175b
Updated
10 months ago
Comments count
1
setup to pyproject
Updated
a year ago
Process blocks when deploying OPT-1.3B with FasterTransformer
Closed
a year ago
train opt-125M from scratch
Updated
a year ago
Comments count
1
How can I pretrain an opt-model with the codes?
Updated
a year ago
pre-training script faild
Closed
a year ago
Comments count
6
Possible feature and bugfix contributions from Microsoft research team's fork of Metaseq
Updated
a year ago
Comments count
4
OPT在中文对话上表现如何呢?
Updated
a year ago
Grammatical Error Correction (GEC) prompt for OPT-IML
Updated
a year ago
load checkpoint failed when training with multi-nodes.
Closed
a year ago
Comments count
1
FSDP is incompatible with BF16
Closed
a year ago
Comments count
4
Sub-workers exits without messages
Updated
a year ago
Comments count
6
OPT and LLaMA
Closed
a year ago
Comments count
1
How to finetune from a consolidated model ?
Updated
a year ago
Comments count
1
Add type hints to all methods
Updated
a year ago
Deprecate convert_to_singleton
Updated
a year ago
Comments count
13
Converting OPT-175B tokenizer to HF format?
Updated
a year ago
Comments count
2
Confirm md5sums after running reshard_fsdp.py on OPT-175B #702
Closed
a year ago
Comments count
3
Incorrect md5sums after running reshard_fsdp.py on OPT-175B
Closed
a year ago
Comments count
2
generation_args["stop"] doesn't work for stop sequence "\n\n"
Updated
a year ago
downloading opt-66B part7 get access denied
Closed
a year ago
Comments count
1
Torch setup
Closed
a year ago
Comments count
2
convert_to_singleton doesn't seem to handle bias properly
Updated
a year ago
`reshard_mp.py --num-output-parts 1` merges to smaller OPT file
Closed
a year ago
Comments count
3
Re-release consolidated OPT / OPT-IML checkpoints
Updated
a year ago
Comments count
2
Implement `finish_reason` in API response
Updated
a year ago
Remove ` --reset-lr-scheduler` flag
Updated
a year ago
Very weird predictions of OPT-IML-30B on Blended Skill Talk dataset.
Closed
a year ago
Comments count
4
OPT-IML Bench release?
Closed
a year ago
Comments count
2
A scholar who loves AI asked
Closed
a year ago
Comments count
1
RuntimeError
Closed
a year ago
Comments count
1
Access request for opt-175b
Closed
a year ago
Comments count
1
Errors:Could not load 'base_config'
Updated
a year ago
Comments count
2
How much gpu memory needed for opt-175B fine-tuning?
Closed
a year ago
Comments count
2
Example scripts for training is not working at all.
Closed
a year ago
Comments count
1
Implement --update-freq for StreamingLanguageModeling task
Updated
a year ago
Convert to singleton script doesn't work using a unified tokenizer file
Closed
a year ago
Bring CM3 in!
Closed
a year ago
how to get sharded ckpt
Updated
a year ago
Comments count
4
No response for access of 175B checkpoints
Closed
a year ago
Failure after loading checkpoint shards.
Closed
a year ago
metaseq-train hangs at global barrier when 2-nodes launch.
Closed
a year ago
Comments count
5
how to launch pretrain jobs when no slurm cluster
Closed
a year ago
kill code of "zero_sharding" in metaseq
Closed
a year ago
How long does it take to get a reply after filling out the Request Form ?
Closed
a year ago
Comments count
1
Why is ReLU used?
Closed
a year ago
Comments count
2
Previous
Next