INDEX

Explanations

numerical values related to measurements of time, quantity, or attributes

oai_token-act-pair · gpt-3.5-turbo

numerical values and quantities related to time or measurements

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

 Elias

-0.73

ãĤ¦

-0.68

icism

-0.67

achu

-0.67

 Parish

-0.66

employment

-0.66

rique

-0.63

 blamed

-0.62

hammad

-0.62

illes

-0.62

POSITIVE LOGITS

%-

1.06

0.96

px

0.94

ÃĹ

0.92

Â°

0.90

 litres

0.88

 watts

0.88

0.85

Activations Density 0.268%

numerical values related to measurements of time, quantity, or attributes

numerical values and quantities related to time or measurements

No Comments

No Known Activations