INDEX

Explanations

mentions of meals

oai_token-act-pair · gpt-3.5-turbo

references to meals and meal planning

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ignty

-0.71

idency

-0.69

inates

-0.67

itars

-0.61

sighted

-0.61

 Morse

-0.60

 Jericho

-0.59

 Personality

-0.59

 bluff

-0.59

 Painter

-0.58

POSITIVE LOGITS

worms

1.27

 meal

1.19

 meals

1.12

 Meal

1.09

 eaten

0.92

worm

0.91

 buffet

0.89

mares

0.88

meat

0.80

winner

0.79

Activations Density 0.009%

mentions of meals

references to meals and meal planning

No Comments

No Known Activations