INDEX

Explanations

sweet-tasting food-related words

oai_token-act-pair · gpt-3.5-turbo

references to sweetness 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 @ 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

Downloadha

-1.20

 Shutterstock

-0.76

senal

-0.76

ctica

-0.76

 Administ

-0.75

kson

-0.72

agnar

-0.72

 Occupations

-0.71

NUM

-0.69

Uz

-0.66

POSITIVE LOGITS

eners

1.37

heart

1.37

ened

1.20

ener

1.16

bread

1.12

sweet

1.08

 potato

1.08

est

1.08

ening

1.04

 sweet

0.98

Activations Density 0.014%

sweet-tasting food-related words

references to sweetness in various contexts

No Comments

No Known Activations