INDEX

Explanations

words related to unethical or immoral behavior

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "dirty" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

*/(

-1.21

PsyNetMessage

-0.85

HCR

-0.80

itech

-0.78

iphate

-0.77

ajor

-0.76

ihilation

-0.75

aic

-0.75

nesota

-0.75

ãĥĦ

-0.74

POSITIVE LOGITS

 tricks

1.22

 laundry

1.17

 linen

0.97

 trick

0.92

 dirty

0.92

 rotten

0.90

 diapers

0.88

 dishes

0.87

 little

0.87

glers

0.86

Activations Density 0.049%

words related to unethical or immoral behavior

references to the concept of "dirty" in various contexts

No Comments

No Known Activations