INDEX

Explanations

words related to legal proceedings or hearings

oai_token-act-pair · gpt-3.5-turbo

references to legal or formal hearings

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

oil

-0.75

jac

-0.75

xon

-0.70

vier

-0.69

spr

-0.68

ecd

-0.68

hidden

-0.66

isine

-0.66

eros

-0.66

obs

-0.64

POSITIVE LOGITS

 hearings

1.15

 Hearing

1.03

 hearing

0.93

 Hear

0.92

imony

0.91

 magistrate

0.79

 testimony

0.78

 examiner

0.77

 Isle

0.76

 Transcript

0.76

Activations Density 0.023%

words related to legal proceedings or hearings

references to legal or formal hearings

No Comments

No Known Activations