INDEX

Explanations

phrases that are likely identifiers of specific individuals

oai_token-act-pair · gpt-3.5-turbo

instances of a specific character or symbol

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 multiplying

-0.71

 guiActiveUnfocused

-0.68

 sweeping

-0.68

 demoral

-0.68

 scattering

-0.67

 versa

-0.66

 shack

-0.65

 scaven

-0.64

mid

-0.64

Veg

-0.63

POSITIVE LOGITS

£

1.08

ı

1.05

¬

0.94

İ

0.92

§

0.92

Ĵ

0.91

¹

0.91

į

0.87

Ń

0.87

¡

0.87

Activations Density 0.126%

phrases that are likely identifiers of specific individuals

instances of a specific character or symbol

No Comments

No Known Activations