INDEX

Explanations

phrases related to legal entities such as district attorneys' offices

oai_token-act-pair · gpt-3.5-turbo

references to legal authorities and their offices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zl

-0.77

-0.73

bull

-0.72

oken

-0.69

Ku

-0.68

advertising

-0.66

Condition

-0.63

pless

-0.63

hend

-0.62

wcsstore

-0.61

POSITIVE LOGITS

 prosecuting

0.83

 investigating

0.82

iate

0.82

 indicted

0.79

 prosecutor

0.78

lege

0.77

 investigation

0.76

 investigator

0.76

ial

0.74

 prosecuted

0.73

Activations Density 0.040%

phrases related to legal entities such as district attorneys' offices

references to legal authorities and their offices

No Comments

No Known Activations