INDEX

Explanations

food ingredients and dishes

oai_token-act-pair · gpt-3.5-turbo

food-related nouns

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

Ability

-0.73

Function

-0.73

Specific

-0.73

Phase

-0.72

":["

-0.72

 Doctrine

-0.72

SPONSORED

-0.70

Jr

-0.70

JM

-0.67

 Offic

-0.66

POSITIVE LOGITS

cake

1.08

bowl

1.02

 popcorn

0.95

 pies

0.93

pipe

0.88

 potato

0.88

cakes

0.87

bag

0.87

vine

0.87

 soup

0.86

Activations Density 0.263%

food ingredients and dishes

food-related nouns

No Comments

No Known Activations