INDEX

Explanations

proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

references to prominent figures in images or captions related to events and contexts

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

 rall

-0.79

 desk

-0.63

 comprom

-0.63

 forwards

-0.63

 erect

-0.62

 quar

-0.62

 pageant

-0.62

 stru

-0.62

 fences

-0.62

 wedd

-0.60

POSITIVE LOGITS

1.21

1.11

Secondly

0.99

Â²

0.99

 Secondly

0.92

 Second

0.86

Second

0.84

II

0.83

0.82

second

0.82

Activations Density 0.125%

proper nouns and names

references to prominent figures in images or captions related to events and contexts

No Comments

No Known Activations