INDEX

Explanations

fruit names

oai_token-act-pair · gpt-3.5-turbo

references to various types of fruits

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

hematic

-0.82

RAW

-0.80

 condem

-0.80

kson

-0.76

ciplinary

-0.73

raq

-0.71

rict

-0.70

iazep

-0.69

earable

-0.68

gregation

-0.68

POSITIVE LOGITS

fruit

1.43

 fruit

1.20

 juice

1.15

 fruits

1.13

 strawberries

1.02

 cherry

1.02

 tomatoes

0.98

 mango

0.97

 strawberry

0.96

 juices

0.95

Activations Density 0.031%

fruit names

references to various types of fruits

No Comments

No Known Activations