INDEX

Explanations

phrases or sentences describing visual appearances or observations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating uncertainty or appearance

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

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

ridges

-0.65

obiles

-0.59

aples

-0.58

anders

-0.57

 lamp

-0.53

ondo

-0.53

 lett

-0.53

 VIDEOS

-0.52

letcher

-0.51

ffect

-0.51

POSITIVE LOGITS

 amounted

0.83

 constitute

0.74

 amounts

0.72

 constitutes

0.70

é¾

0.68

kus

0.66

 transpired

0.66

ulously

0.66

imately

0.65

uably

0.63

Activations Density 0.090%

phrases or sentences describing visual appearances or observations

phrases indicating uncertainty or appearance

No Comments

No Known Activations