issues
search
mosaicml
/
examples
Fast and flexible reference benchmarks
Apache License 2.0
441
stars
125
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Seq2Seq finetuning
#304
alextrott16
closed
1 year ago
1
TransformerEngine support in MosaicGPT
#303
dskhudia
closed
1 year ago
1
add multi-query attn
#302
vchiley
closed
1 year ago
1
Add example code to download huggingface checkpoint from OCI for generate
#301
sashaDoubov
closed
1 year ago
0
eval first except when resuming
#300
vchiley
closed
1 year ago
0
updt SchedGC to include cuda.empty_cache
#299
vchiley
closed
1 year ago
0
add gc collect to sgc
#298
vchiley
closed
1 year ago
1
add gc.collect to sgc
#297
vchiley
closed
1 year ago
0
HERO run branch
#296
vchiley
closed
1 year ago
1
add sched gc
#295
vchiley
closed
1 year ago
0
Grammar check courtesy of Emily
#294
vchiley
closed
1 year ago
0
Merge release 0.0.4 back to main
#293
dakinggg
closed
1 year ago
0
1 out of N runs starts successfully, others fail immediately
#292
eldarkurtic
closed
1 year ago
9
Fix generation callback interval
#291
dakinggg
closed
1 year ago
0
Update README, add low precision option for `convert_composer_to_hf.py`
#290
abhi-mosaic
closed
1 year ago
0
Update yamls
#289
dakinggg
closed
1 year ago
0
Add Monolithic Checkpoint Callback
#288
abhi-mosaic
closed
1 year ago
1
Monolithic Checkpoint Callback
#287
eracah
closed
1 year ago
1
Pass in `save_latest_filename`
#286
abhi-mosaic
closed
1 year ago
0
enable user to pass in save_filename
#285
vchiley
closed
1 year ago
0
Revert "make LlamaTokenizer work for tokenizer bakeoff"
#284
samhavens
closed
1 year ago
0
Updates the export for onnx script to work with HF models
#283
dakinggg
closed
1 year ago
2
make LlamaTokenizer work for tokenizer bakeoff
#282
samhavens
closed
1 year ago
5
make LlamaTokenizer work for tokenizer bakeoff
#281
samhavens
closed
1 year ago
0
Improve bin packing efficiency
#280
alextrott16
closed
1 year ago
0
Update mcloud_run.yaml
#279
A-Jacobson
closed
1 year ago
0
Add `runtime_estimator`, remove `optimization_level`
#278
abhi-mosaic
closed
1 year ago
0
Add HF conversion and generate scripts
#277
abhi-mosaic
closed
1 year ago
1
updt tests
#276
vchiley
closed
1 year ago
0
updt dl
#275
vchiley
closed
1 year ago
1
Dreambooth example
#274
A-Jacobson
closed
1 year ago
0
Upgrade to Composer 0.13.3
#273
abhi-mosaic
closed
1 year ago
0
Make MosaicGPT MFU calculator more robust
#272
abhi-mosaic
closed
1 year ago
0
updt throughput tables
#271
vchiley
closed
1 year ago
0
Update training YAMLs with new defaults
#270
abhi-mosaic
closed
1 year ago
2
Add a trlx example
#269
dakinggg
closed
1 year ago
0
Add bin packing collator wrapper in denoising
#268
alextrott16
closed
1 year ago
1
Inference benchmarking
#267
bcui19
closed
1 year ago
2
Support only-within-sequence attention for MosaicGPT
#266
alextrott16
closed
1 year ago
2
Add a callback that logs generations to wandb at eval end
#265
dakinggg
closed
1 year ago
2
set cache default to true for generation
#264
dakinggg
closed
1 year ago
0
inf and use_cache into release
#263
vchiley
closed
1 year ago
3
updt flg
#262
vchiley
closed
1 year ago
1
Use HF generate for inference
#261
dskhudia
closed
1 year ago
0
Model builder fn err updt
#260
vchiley
closed
1 year ago
0
fix bias
#259
vchiley
closed
1 year ago
0
fix model_max_length setting
#258
dakinggg
closed
1 year ago
0
Fix slicing for padding + cache
#257
dakinggg
closed
1 year ago
0
Add support for output hidden states (for contrastive decoding) and beam search with past
#256
dakinggg
closed
1 year ago
0
Add torch 2.0 based tensor parallel support
#255
dskhudia
closed
1 year ago
6
Previous
Next