INDEX

Explanations

phrases that involve looking or observing

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

phrases about observation or examination

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

uala

-0.61

 orally

-0.59

 facult

-0.59

ilts

-0.58

accompan

-0.57

 restraining

-0.57

learning

-0.57

 verbally

-0.56

ciating

-0.56

aten

-0.56

POSITIVE LOGITS

WARD

0.81

oward

0.73

ificant

0.71

 Compass

0.69

asma

0.67

 suspicious

0.66

ivable

0.66

 noses

0.65

 IMAGES

0.65

obin

0.64

Activations Density 0.095%

phrases that involve looking or observing

phrases about observation or examination

No Comments

No Known Activations