INDEX

Explanations

phrases related to reciprocation or actions/events mirroring one another

oai_token-act-pair · gpt-3.5-turbo

references to mutual relationships or exchanges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gotten

-0.64

gravity

-0.61

ohyd

-0.60

Bang

-0.59

fly

-0.58

stra

-0.57

utters

-0.57

 synopsis

-0.57

aquin

-0.56

 Presence

-0.56

POSITIVE LOGITS

 versa

1.16

worldly

0.78

forth

0.74

é»Ĵ

0.72

forward

0.72

isSpecialOrderable

0.70

dayName

0.70

ipop

0.70

 parted

0.68

entimes

0.65

Activations Density 0.011%

phrases related to reciprocation or actions/events mirroring one another

references to mutual relationships or exchanges

No Comments

No Known Activations