INDEX

Explanations

phrases related to transformation or change

oai_token-act-pair · gpt-3.5-turbo

phrases indicating changes or transformations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ollen

-0.77

complex

-0.68

ogle

-0.67

times

-0.65

fact

-0.63

fal

-0.63

bats

-0.63

anwhile

-0.63

pack

-0.62

issues

-0.61

POSITIVE LOGITS

 Archdemon

0.75

 Reincarnated

0.73

ezvous

0.69

livion

0.66

ragon

0.63

igslist

0.63

 Disapp

0.62

 TAMADRA

0.61

amiya

0.61

 onto

0.59

Activations Density 0.172%

phrases related to transformation or change

phrases indicating changes or transformations

No Comments

No Known Activations