INDEX

Explanations

references to internal matters or investigations

oai_token-act-pair · gpt-3.5-turbo

references to internal processes or investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

vous

-0.87

oÄŁ

-0.84

apo

-0.83

eful

-0.83

ky

-0.78

gger

-0.74

some

-0.74

eday

-0.72

ammy

-0.72

kers

-0.71

POSITIVE LOGITS

 combustion

1.18

 workings

1.11

 organs

0.94

ized

0.93

ization

0.93

 affairs

0.91

 deliberations

0.79

ised

0.76

izing

0.76

 Internal

0.74

Activations Density 0.012%

references to internal matters or investigations

references to internal processes or investigations

No Comments

No Known Activations