INDEX

Explanations

references to rules, guidelines, or regulations

oai_token-act-pair · gpt-3.5-turbo

mentions of various policies

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

htaking

-0.88

ITNESS

-0.85

eneg

-0.80

Ago

-0.77

 Flavoring

-0.77

bane

-0.75

 Stain

-0.74

pine

-0.73

alsh

-0.71

semble

-0.71

POSITIVE LOGITS

 policy

0.97

 enforcement

0.94

holders

0.86

 policies

0.86

 Enforcement

0.86

 enforced

0.82

 stance

0.81

 prohibiting

0.81

 discrim

0.80

policy

0.79

Activations Density 0.024%

references to rules, guidelines, or regulations

mentions of various policies

No Comments

No Known Activations