INDEX

Explanations

phrases related to legal allegations

oai_token-act-pair · gpt-3.5-turbo

instances of claims or accusations being made

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

cffffcc

-0.86

ammy

-0.74

heartedly

-0.65

 screening

-0.64

outing

-0.63

 spot

-0.61

talk

-0.60

 Unch

-0.59

outed

-0.57

 captcha

-0.55

POSITIVE LOGITS

ments

0.89

 alleges

0.87

reau

0.83

lements

0.78

MENT

0.76

zynski

0.76

TAIN

0.75

cing

0.75

 allege

0.74

olate

0.73

Activations Density 0.048%

phrases related to legal allegations

instances of claims or accusations being made

No Comments

No Known Activations