issues
search
microsoft
/
TransformerCompression
For releasing code related to compression methods for transformers, accompanying our publications
MIT License
344
stars
30
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Quarot: DeepSeek-V2 Support
#174
RanchiZhao
opened
2 weeks ago
0
How to evaluate the sliced model
#173
1250826219
opened
2 weeks ago
0
Update README and latest from master
#172
nailimixaM
closed
1 month ago
0
Add Mixtral and better groupwise quantization
#171
nailimixaM
closed
1 month ago
0
James/quantiles
#170
jameshensman
closed
1 month ago
0
James/activation grouping
#169
jameshensman
closed
1 month ago
0
James/scales
#168
jameshensman
closed
1 month ago
0
How to finetune with multi-gpus under data parallel setting?
#167
kriskrisliu
opened
1 month ago
0
initial attempt at grouping
#166
jameshensman
closed
1 month ago
0
What is the number of parameters afrer slicing?
#165
yaya-sy
closed
1 month ago
2
PHI包导入问题
#164
1250826219
closed
1 month ago
2
James/gptq improvements
#163
jameshensman
closed
1 month ago
0
Add Phi3
#162
nailimixaM
closed
1 month ago
0
Phi3 fixes
#161
nailimixaM
closed
2 months ago
0
model inference
#160
ChrisXULC
opened
2 months ago
0
Add GPTQ into Act Quant Branch
#159
nailimixaM
closed
2 months ago
0
Add GPTQ
#158
nailimixaM
closed
2 months ago
0
Add GPTQ
#157
nailimixaM
closed
2 months ago
0
QuaRot bugfixes
#156
nailimixaM
closed
2 months ago
0
error when Fine-tuning a sliced model llama 3
#155
ChrisXULC
closed
2 months ago
0
Add phi3 quarot
#154
pashminacameron
closed
1 month ago
0
Remove fast hadamard transform dependency. Use scipy ref impl
#153
pashminacameron
closed
1 month ago
0
Add abstraction for QuarotFP16Linear layers in RTN quantization
#152
pashminacameron
closed
1 month ago
0
Update llama repo emb interface for newer transformers
#151
pashminacameron
closed
2 months ago
0
Update transformers to 4.41.0
#150
pashminacameron
closed
2 months ago
3
QuaRot: Add activation and KV cache quantization, GPTQ, Phi3, Groupsizes
#149
nailimixaM
closed
1 month ago
0
QuaRot: KV cache quantization
#148
nailimixaM
closed
2 months ago
0
Add Llama3 support to llama_adapter
#147
radhikamp99
closed
2 months ago
5
Use a task metric map in lm_eval runner
#146
pashminacameron
closed
2 months ago
0
Add Phi-3-mini adapter
#145
pashminacameron
closed
2 months ago
0
Update dependencies
#144
msdmkats
closed
2 months ago
1
QuaRot: cascade into quarot main
#143
nailimixaM
closed
1 month ago
6
Add QuaRot (no quantization yet)
#142
nailimixaM
closed
2 months ago
2
A problem about the PPL value after sliced model fine-tuning
#141
qxpBlog
closed
1 month ago
2
How to load the tuned slicemodel
#140
qxpBlog
closed
1 month ago
1
Make sliced models HuggingFace compatible
#139
LianaMikael
opened
3 months ago
3
NotImplementedError: xx is neither a Hugging Face model nor a supported local model.
#138
qxpBlog
closed
3 months ago
2
Refactor QuaRot [WIP]
#137
nailimixaM
closed
3 months ago
1
Bump transformers from 4.37 to 4.38.0
#136
dependabot[bot]
closed
2 months ago
1
Update transformers and lm-eval
#135
pashminacameron
closed
3 months ago
0
Remove monkeypatch from QuaRot source
#134
nailimixaM
closed
3 months ago
1
Command R and R+ support
#133
Steel-skull
opened
3 months ago
1
Fix phi2 adapter test
#132
nailimixaM
closed
3 months ago
0
Question of `RMSNorm`'s `forward` function
#131
zhaoyang-star
closed
3 months ago
2
Add QuaRot (RTN) [WIP]
#130
nailimixaM
closed
2 months ago
1
Layer fusion with Llama
#129
kiucho
closed
3 months ago
2
How to run with llama-1?
#128
liuxiaozhu01
closed
3 months ago
4
can't reproduce result
#127
MrGGLS
closed
3 months ago
8
can't install slicegpt using "pip install -e." on CPU platform
#126
JCDemon
closed
1 month ago
2
Support sliced local model loading from path
#125
pashminacameron
closed
3 months ago
2
Next