INDEX

Explanations

phrases or words indicating a state of comparison or relation between different entities

oai_token-act-pair · gpt-3.5-turbo

terms indicating inclusion or being part of a group

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

enders

-0.72

irs

-0.70

iens

-0.67

end

-0.65

eri

-0.65

ending

-0.65

tto

-0.65

dim

-0.65

earable

-0.64

 Reloaded

-0.64

POSITIVE LOGITS

 Whilst

1.00

 whilst

0.88

ĨĴ

0.82

sembly

0.79

CLASSIFIED

0.77

ĵĺ

0.77

 obser

0.75

alogue

0.74

Ħ¢

0.72

abama

0.70

Activations Density 0.011%

phrases or words indicating a state of comparison or relation between different entities

terms indicating inclusion or being part of a group

No Comments

No Known Activations