INDEX

Explanations

reporting speech or stating information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating reported speech or quotations

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

Lens

-0.75

actly

-0.72

Hung

-0.71

Pont

-0.70

Enlarge

-0.66

pel

-0.65

kay

-0.64

joice

-0.63

à¦

-0.62

ãĥİ

-0.61

POSITIVE LOGITS

 moreover

0.93

 furthermore

0.91

 secondly

0.81

 afterward

0.72

 afterwards

0.71

:"

0.69

 therefore

0.67

 bluntly

0.66

imaru

0.66

"[

0.64

Activations Density 0.168%

reporting speech or stating information

phrases indicating reported speech or quotations

No Comments

No Known Activations