INDEX

Explanations

medical terms or conditions related to genetics

oai_token-act-pair · gpt-3.5-turbo

terms associated with deception or dishonesty

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

unci

-0.76

ested

-0.72

 Skydragon

-0.70

 Kenobi

-0.69

asking

-0.68

unciation

-0.68

stadt

-0.67

DERR

-0.67

icip

-0.67

caster

-0.67

POSITIVE LOGITS

utical

1.01

ce

0.99

les

0.90

pter

0.89

rette

0.85

llan

0.78

re

0.77

0.76

lled

0.76

ment

0.75

Activations Density 0.021%

medical terms or conditions related to genetics

terms associated with deception or dishonesty

No Comments

No Known Activations