INDEX

Explanations

mentions of legal terms and courtroom proceedings

oai_token-act-pair · gpt-3.5-turbo

references to defendants in a legal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ories

-0.83

yip

-0.80

Ã¼r

-0.80

olen

-0.77

atism

-0.77

eful

-0.75

[|

-0.74

efully

-0.72

UNCH

-0.71

unch

-0.70

POSITIVE LOGITS

 pled

0.84

 defendants

0.78

 defendant

0.75

 plead

0.74

 Defendant

0.72

 accused

0.68

 sentenced

0.68

 convicted

0.68

 contracted

0.66

 briefs

0.66

Activations Density 0.014%

mentions of legal terms and courtroom proceedings

references to defendants in a legal context

No Comments

No Known Activations