INDEX

Explanations

business-related terms and locations

oai_token-act-pair · gpt-3.5-turbo

nouns related to places, institutions, and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ogenous

-0.75

asonable

-0.75

urate

-0.73

 Helpful

-0.69

usually

-0.67

à¨

-0.66

opping

-0.65

abolic

-0.65

asive

-0.64

ificant

-0.64

POSITIVE LOGITS

 detachment

0.70

 Ajax

0.65

 consortium

0.63

Cyn

0.63

 Etsy

0.62

 suprem

0.60

 pictured

0.59

HMS

0.58

soDeliveryDate

0.58

 championed

0.58

Activations Density 0.632%

business-related terms and locations

nouns related to places, institutions, and entities

No Comments

No Known Activations