INDEX

Explanations

phrases related to current time or the present day

oai_token-act-pair · gpt-3.5-turbo

references to contemporary issues and the present time

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

henko

-0.77

-0.73

afort

-0.68

sym

-0.67

bid

-0.66

Sym

-0.66

essee

-0.65

meat

-0.63

alcohol

-0.62

med

-0.61

POSITIVE LOGITS

adays

0.99

 morning

0.88

abouts

0.85

 afternoon

0.85

days

0.83

 evening

0.74

 Today

0.72

 mornings

0.67

JD

0.64

 fashionable

0.64

Activations Density 0.040%

phrases related to current time or the present day

references to contemporary issues and the present time

No Comments

No Known Activations