INDEX

Explanations

positive or neutral adjectives related to cleanliness, safety, or harmlessness

oai_token-act-pair · gpt-3.5-turbo

terms related to safety and mildness

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

KER

-0.75

illon

-0.67

hon

-0.64

amp

-0.63

Sharp

-0.62

hod

-0.62

AMP

-0.62

lined

-0.61

aeda

-0.61

obook

-0.61

POSITIVE LOGITS

 harmless

1.09

 benign

0.91

ality

0.88

 bystand

0.86

mate

0.78

alities

0.78

 innocuous

0.77

»Ĵ

0.75

alty

0.72

ishable

0.68

Activations Density 0.016%

positive or neutral adjectives related to cleanliness, safety, or harmlessness

terms related to safety and mildness

No Comments

No Known Activations