INDEX

Explanations

phrases related to criminal activities and investigations

oai_token-act-pair · gpt-3.5-turbo

actions and phrases related to commands or requests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

iasm

-0.83

alg

-0.82

wayne

-0.71

gae

-0.68

til

-0.68

aho

-0.67

anse

-0.64

enment

-0.64

alach

-0.64

tions

-0.62

POSITIVE LOGITS

 inappropriately

1.11

 prostitutes

0.92

 inappropriate

0.90

 unlawfully

0.88

 improper

0.88

 improperly

0.88

 illegally

0.88

 bribes

0.84

 deliberately

0.83

 objectionable

0.79

Activations Density 0.460%

phrases related to criminal activities and investigations

actions and phrases related to commands or requests

No Comments

No Known Activations