INDEX

Explanations

phrases indicating absence or lack of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating safety or lack of harm

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

auga

-0.79

ivities

-0.75

ById

-0.70

ilated

-0.66

listed

-0.64

hack

-0.62

interstitial

-0.61

enstein

-0.61

bra

-0.60

las

-0.60

POSITIVE LOGITS

nor

0.94

except

0.91

Except

0.90

VPN

0.81

Moreover

0.81

 except

0.81

Instead

0.80

}"

0.79

 furthermore

0.78

Likewise

0.78

Activations Density 0.726%

phrases indicating absence or lack of something

phrases indicating safety or lack of harm

No Comments

No Known Activations