INDEX

Explanations

instances where someone is unaware of specific information

oai_token-act-pair · gpt-3.5-turbo

instances where people are unaware of important information or facts

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

ioxide

-0.78

ŃĶ

-0.71

attery

-0.69

ione

-0.66

oulos

-0.65

 exting

-0.64

igl

-0.63

rak

-0.62

ischer

-0.62

 earthqu

-0.62

POSITIVE LOGITS

 anymore

1.19

 anything

1.15

how

1.08

 what

1.05

nor

1.03

why

0.98

 whatsoever

0.98

 whether

0.97

 whats

0.95

WHY

0.93

Activations Density 0.100%

instances where someone is unaware of specific information

instances where people are unaware of important information or facts

No Comments

No Known Activations