INDEX

Explanations

introductory phrases signaling the beginning of a discussion or explanation

oai_token-act-pair · gpt-3.5-turbo

introductory phrases and expressions of intent or preparation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĥ´

-0.69

Wed

-0.67

ãĤ´ãĥ³

-0.65

ãĥĩ

-0.64

 Niet

-0.60

 Pear

-0.60

ardless

-0.58

Fax

-0.56

yles

-0.56

Wem

-0.53

POSITIVE LOGITS

 briefly

0.95

 suffice

0.90

 acquaint

0.90

Firstly

0.87

 clarify

0.83

 basics

0.79

 realise

0.77

 disclaimer

0.76

 realize

0.76

 remind

0.75

Activations Density 0.122%

introductory phrases signaling the beginning of a discussion or explanation

introductory phrases and expressions of intent or preparation

No Comments

No Known Activations