INDEX

Explanations

mentions of television series

oai_token-act-pair · gpt-3.5-turbo

mentions of television series

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

zie

-0.86

orget

-0.85

attled

-0.79

artisan

-0.79

imentary

-0.72

lesi

-0.72

lig

-0.71

Vie

-0.70

nai

-0.70

arest

-0.70

POSITIVE LOGITS

 finale

0.92

 Reborn

0.80

 regulars

0.75

 adaptation

0.74

manship

0.74

enegger

0.72

 iteration

0.71

 Reign

0.67

 incarnation

0.65

runners

0.65

Activations Density 0.022%

mentions of television series

mentions of television series

No Comments

No Known Activations