INDEX

Explanations

mentions of government entities and officials

oai_token-act-pair · gpt-3.5-turbo

references to specific entities or offices

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

 Valiant

-0.83

gib

-0.73

 clich

-0.71

 sacrific

-0.71

 scattering

-0.70

Veg

-0.69

 delusion

-0.69

 Babel

-0.69

 Golem

-0.68

 Mermaid

-0.68

POSITIVE LOGITS

âĢł

0.93

¹

0.92

£

0.89

ISIS

0.87

âģ

0.82

taboola

0.82

Colorado

0.82

Solution

0.80

ı

0.79

sure

0.78

Activations Density 0.593%

mentions of government entities and officials

references to specific entities or offices

No Comments

No Known Activations