INDEX

Explanations

deep-fried food items

oai_token-act-pair · gpt-3.5-turbo

terms related to fried foods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

QB

-0.82

Church

-0.68

Org

-0.66

Newsletter

-0.66

atican

-0.65

qq

-0.65

xus

-0.64

 Visitors

-0.63

 Volume

-0.63

orsi

-0.63

POSITIVE LOGITS

 fried

1.54

 frying

1.32

fry

1.28

 fries

1.08

 cakes

1.06

 tofu

1.06

 crispy

1.06

 dough

1.01

 toast

1.00

fried

0.97

Activations Density 0.009%

deep-fried food items

terms related to fried foods

No Comments

No Known Activations