INDEX

Explanations

words related to changes or variations in quantity

oai_token-act-pair · gpt-3.5-turbo

words and phrases that indicate an increase or decrease in something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

mad

-0.79

ilo

-0.72

spring

-0.65

DragonMagazine

-0.64

tta

-0.64

ey

-0.64

gins

-0.64

æ³

-0.61

Tok

-0.61

Â©

-0.61

POSITIVE LOGITS

 productivity

0.97

 visibility

0.97

 efficiency

0.96

 exponentially

0.94

 perceptions

0.93

 longevity

0.92

 morale

0.90

 tremend

0.89

 perception

0.87

 tremendously

0.86

Activations Density 0.126%

words related to changes or variations in quantity

words and phrases that indicate an increase or decrease in something

No Comments

No Known Activations