INDEX

Explanations

references to specific television shows related to law and order

oai_token-act-pair · gpt-3.5-turbo

references to the legal system and law enforcement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

auga

-1.00

hesda

-0.86

iliary

-0.85

ngth

-0.78

otos

-0.72

riks

-0.71

yip

-0.70

ovie

-0.66

hiba

-0.66

ritch

-0.65

POSITIVE LOGITS

enforcement

1.04

abiding

0.94

×¨

0.76

riage

0.76

clinton

0.68

 precedent

0.66

 Veronica

0.65

à¨

0.65

ASED

0.64

 ACTIONS

0.64

Activations Density 0.286%

references to specific television shows related to law and order

references to the legal system and law enforcement

No Comments

No Known Activations