INDEX

Explanations

words related to destruction and negative consequences

oai_token-act-pair · gpt-3.5-turbo

words related to destruction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

rouse

-0.62

icip

-0.61

hey

-0.60

yah

-0.60

 Wilde

-0.60

 voicing

-0.59

 Founders

-0.58

çīĪ

-0.58

 Kislyak

-0.58

uana

-0.57

POSITIVE LOGITS

 havoc

1.04

 wrought

0.92

roying

0.88

 derby

0.84

wre

0.81

adoes

0.79

hower

0.76

struction

0.75

ados

0.73

 inflicted

0.73

Activations Density 0.141%

words related to destruction and negative consequences

words related to destruction

No Comments

No Known Activations