Giters
FMInference
/
DejaVu
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
250
Watchers:
6
Issues:
29
Forks:
31
FMInference/DejaVu Issues
How to run the code with smaller ckpt like opt-6.7B
Updated
24 days ago
Comments count
10
the issue of parameter settings
Updated
a month ago
Some questions about implementation details
Updated
2 months ago
Comments count
3
Failed to collect training data for sparsity predictor
Updated
2 months ago
Comments count
8
How to run on a single GPU?
Updated
2 months ago
Comments count
3
Questions about Attention Sparsity Implementation
Closed
3 months ago
PyTorch 1.12 and flash-attn==0.2.8 are not compatible.
Updated
3 months ago
the code about collect activations while inferencing
Updated
3 months ago
Miss csrc folder when building docker images
Updated
3 months ago
Comments count
3
Predictor without activation function?
Closed
a year ago
Comments count
3
No output in collecting training data
Closed
10 months ago
Comments count
13
A question about attention block sparsity
Updated
4 months ago
Clarification on Output Neuron Pruning Method in "Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time"
Closed
4 months ago
Comments count
1
In fp16, the sparse kernel is slower than PyTorch dense gemm
Closed
4 months ago
1
Closed
5 months ago
Does anyone know if this work is implemented on llama?
Updated
5 months ago
Comments count
3
Please give a instruction to run Hardware-efficient Implementation dejaVu model inferenc,I Can't find the moudle src.models.gpt_sparse
Updated
6 months ago
Questions on sparse MLP implementation
Closed
a year ago
Comments count
3
Question about OPT-30B
Updated
7 months ago
Training MLP layer Issue
Updated
7 months ago
Comments count
1
Missing full.pt when running latency benchmark
Updated
7 months ago
LLama 2 compatability
Updated
8 months ago
Compare the sparse model on downstream tasks
Updated
10 months ago
Comments count
2
Missing lm_eval harness for downstream task prediction
Updated
10 months ago
Inconsisten File Names
Updated
10 months ago
Missing Argument and Filtering
Updated
10 months ago
fix file name format as they are named like *_sp_x_*
Updated
10 months ago
AttributeError: 'OPTAttention' object has no attribute 'fp_query'
Closed
10 months ago
Comments count
5