INDEX

Explanations

food-related terms

oai_token-act-pair · gpt-3.5-turbo

references to food

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

 Wink

-0.88

 Freem

-0.72

-0.71

 Caldwell

-0.71

Jed

-0.70

orp

-0.69

Gib

-0.69

 Painter

-0.67

aird

-0.67

assis

-0.67

POSITIVE LOGITS

stuff

1.16

 eaten

1.01

grain

1.01

borne

1.00

corn

0.96

 tasted

0.93

 allergies

0.92

 poisoning

0.90

 starch

0.90

 additives

0.88

Activations Density 0.035%

food-related terms

references to food

No Comments

No Known Activations