INDEX

Explanations

legal and judicial terms and proceedings

oai_token-act-pair · gpt-3.5-turbo

references to legal arguments or notable court cases

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Others

-0.77

 others

-0.70

Others

-0.68

"],"

-0.63

ktop

-0.58

adv

-0.58

catentry

-0.58

ellery

-0.58

ibliography

-0.57

 counselling

-0.57

POSITIVE LOGITS

 shockingly

1.14

 ostensibly

1.08

 inexpl

1.04

 horrend

0.99

 egreg

0.97

 supposedly

0.93

 ridiculously

0.90

 goddamn

0.90

 insanely

0.90

 predictably

0.89

Activations Density 1.443%

legal and judicial terms and proceedings

references to legal arguments or notable court cases

No Comments

No Known Activations