INDEX

Explanations

proper nouns related to financial institutions or organizations

oai_token-act-pair · gpt-3.5-turbo

references to countries and regions

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

bern

-0.70

ãĥ´

-0.70

alist

-0.69

inant

-0.66

andem

-0.65

Ð¼

-0.64

comings

-0.62

lying

-0.61

jad

-0.60

Ì

-0.59

POSITIVE LOGITS

 Holdings

0.77

Sharp

0.76

 Publishing

0.74

 Corp

0.71

Ltd

0.70

 Laboratories

0.69

 Foundation

0.67

 Entertainment

0.67

opa

0.66

 Consulting

0.64

Activations Density 0.138%

proper nouns related to financial institutions or organizations

references to countries and regions

No Comments

No Known Activations