INDEX

Explanations

references to the word "Stone"

oai_token-act-pair · gpt-3.5-turbo

the frequent mentions of "Stone"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

olulu

-0.97

merce

-0.88

ntil

-0.79

unct

-0.77

oresc

-0.77

uates

-0.76

ornia

-0.75

uate

-0.74

unal

-0.73

 unrestricted

-0.72

POSITIVE LOGITS

Stone

1.06

hill

1.02

 Stone

0.97

breaker

0.95

bats

0.88

hook

0.88

fish

0.87

falls

0.86

rock

0.86

works

0.83

Activations Density 0.006%

references to the word "Stone"

the frequent mentions of "Stone"

No Comments

No Known Activations