INDEX

Explanations

mentions of a TV series

oai_token-act-pair · gpt-3.5-turbo

references to different television series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

-0.78

prus

-0.72

selves

-0.72

emort

-0.70

zan

-0.69

ez

-0.69

aths

-0.68

orgetown

-0.68

zh

-0.68

bell

-0.67

POSITIVE LOGITS

 finale

1.10

manship

0.95

 regulars

0.83

0.76

 premiere

0.76

hips

0.74

ĸļ

0.72

peak

0.72

hip

0.71

 Reborn

0.71

Activations Density 0.044%

mentions of a TV series

references to different television series

No Comments

No Known Activations