INDEX

Explanations

adjectives and verbs related to consuming food or drink

oai_token-act-pair · gpt-3.5-turbo

references to food and eating habits

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eele

-1.04

soDeliveryDate

-0.86

mble

-0.78

yrinth

-0.77

yssey

-0.77

 insur

-0.73

escription

-0.72

Technical

-0.72

CLASS

-0.72

idth

-0.71

POSITIVE LOGITS

 grapes

1.10

 berries

1.03

 grape

1.00

 contaminated

0.97

 cabbage

0.95

 beverages

0.95

 poisoned

0.92

 meals

0.91

 juice

0.90

 vegetables

0.89

Activations Density 0.283%

adjectives and verbs related to consuming food or drink

references to food and eating habits

No Comments

No Known Activations