INDEX

Explanations

mentions of the name "Ber" or variations of it

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "ber" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ENDED

-0.78

URA

-0.71

qqa

-0.69

 respect

-0.66

ENSE

-0.66

-0.64

ãħĭ

-0.63

 dependence

-0.62

BILITIES

-0.62

 pity

-0.61

POSITIVE LOGITS

culosis

1.18

ber

1.04

keley

1.03

wolves

0.94

geon

0.93

deen

0.92

netic

0.91

ding

0.90

lin

0.88

wald

0.86

Activations Density 0.007%

mentions of the name "Ber" or variations of it

the presence of the word "ber" in various contexts

No Comments

No Known Activations