issues
search
jalammar
/
ecco
Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations directly in Jupyter notebooks explaining the behavior of Transformer-based language models (like GPT2, BERT, RoBERTA, T5, and T0).
https://ecco.readthedocs.io
BSD 3-Clause "New" or "Revised" License
1.98k
stars
169
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Can ecco be applied to tabular pytorch models?
#117
eyal-trabelsi
opened
3 months ago
0
feat(LM): attributions for SequenceClassification
#116
jamesjin0516
opened
4 months ago
0
Installing Ecco, unable to install in Colab
#115
SalvatoreRa
opened
6 months ago
3
baseline for sentence classification task using IG/Deeplift?
#114
lkqnaruto
opened
7 months ago
0
Decoder hidden states not found
#113
herbert-zhou
opened
9 months ago
3
Update requirements: fix failing test
#112
SumanthRH
opened
10 months ago
1
installing ecco
#111
shahramzs
opened
10 months ago
2
update requirements and add float16 support
#110
SumanthRH
closed
10 months ago
3
How to serialize the LMOutput object?
#109
JamesCao2048
opened
11 months ago
0
Make dependencies more loose
#108
joaonadkarni
closed
3 months ago
1
Support for pre-generated output
#107
ThomasDelteil
opened
11 months ago
0
How to retrieve salience of some specific words?
#105
CarhoJohn
opened
1 year ago
1
How to add my own model like chinese-bert-wwm-ext
#104
Myoontyee
opened
1 year ago
0
Fix error when calling _prepare_decoder_input_ids_for_generation
#103
SSamDav
closed
1 year ago
0
Google Colab support only Python 3.10
#102
Rolkarolka
opened
1 year ago
7
Install ecco on Apple M1 Max clip fails
#101
verazuo
closed
1 year ago
1
Two more patches
#100
BiEchi
opened
1 year ago
1
Update lm.py to suit OPT-like models
#99
BiEchi
closed
1 year ago
0
Support to LLM like LLAMA-2 and Vicuna?
#98
BiEchi
closed
1 year ago
15
Error encountered when comparing activation matrices in batch generation loop
#97
miguel-kjh
opened
1 year ago
0
I can't find jupyter notebook for Input_Saliency.
#96
SuperBianC
opened
1 year ago
0
Updated dependencies (no changes needed)
#95
rockwalrus
opened
1 year ago
0
Adding GENRE model to ecco
#94
sky-2002
opened
1 year ago
1
Request for feature addition: Visualization of Neuron Activations and Clustered Neuron Firings
#93
cristianestojeda
opened
1 year ago
1
lm.generate and HuggingFace's generate give different results with do_sample=False
#92
noeliaferruz
closed
1 year ago
0
bug -object has no attribute 'lm_head'-
#91
LeonardRanaldi
opened
1 year ago
2
Update model-config.yaml
#90
LeonardRanaldi
closed
1 year ago
5
llama support
#89
srikanthmalla
opened
1 year ago
4
import ecco ERRO
#88
nikipi
opened
1 year ago
9
Presence of character Ġ before each token in output
#87
osmanii
opened
1 year ago
2
AttributeError: 'OutputSeq' object has no attribute 'saliency'
#86
Claus1
opened
1 year ago
4
Bug: Broken links in docs
#85
keyboardAnt
opened
2 years ago
0
Rankings_watch displaying wrong sequence
#84
MiriUll
opened
2 years ago
1
Support for Distributing Model on Multiple GPU's
#81
chrispan68
closed
2 years ago
3
Save vectorized figure
#80
jadermcs
closed
2 years ago
1
Problem with saliency()
#78
Poul-Andersson
closed
2 years ago
1
Support for Constrained Beam Search?
#77
pablogranolabar
closed
2 years ago
1
Update __init__.py
#76
farhanferoz
closed
2 years ago
0
IndexError when using analysis.pwcca
#75
demkejon001
closed
2 years ago
1
How to use ECCO to explain FineTuned T5 for Conditional Generation ?
#74
tejas20
closed
2 years ago
3
token prefix in roberta model?
#73
sarthusarth
closed
2 years ago
4
Running Eccomap for Pre Trained BertForMaskedLM
#70
iamakshay1
opened
2 years ago
1
Remove `tokenizer_config` usage from the library
#67
jalammar
opened
2 years ago
0
KeyError: 'tokenizer_config'
#66
guustfranssensEY
closed
2 years ago
2
Tokenizer has partial token suffix instead of prefix
#65
guustfranssensEY
opened
2 years ago
1
Add Primary Attribution support for BERT classification
#64
jalammar
opened
2 years ago
4
Add support for PEGASUS model
#63
thomas-chong
closed
2 years ago
6
[bugfix] fix comparison of transformers version in lm generate
#62
joaonadkarni
closed
2 years ago
1
[bug fix] seems the hot fix for T5 generation is not need for later version of transformers
#61
litanlitudan
closed
2 years ago
2
Hotfix jan 2022 1
#60
jalammar
closed
2 years ago
0
Next