INDEX

Explanations

numbers or measurements

oai_token-act-pair · gpt-3.5-turbo

characters or symbols related to specific categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

oids

-0.73

 fodder

-0.71

 Citiz

-0.66

 synerg

-0.66

iewicz

-0.63

 avenues

-0.62

 links

-0.62

iera

-0.62

 detail

-0.62

 Vaugh

-0.62

POSITIVE LOGITS

ï¸ı

1.85

ï¸

1.47

ÃĽ

0.95

Â£

0.94

ternity

0.94

STEM

0.89

Â¯

0.88

¹

0.84

taboola

0.84

âķĲâķĲ

0.84

Activations Density 0.039%

numbers or measurements

characters or symbols related to specific categories or classifications

No Comments

No Known Activations