INDEX

Explanations

statements emphasizing or discussing responsibilities

oai_token-act-pair · gpt-3.5-turbo

discussions about personal and social responsibilities

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

vae

-0.84

arthed

-0.76

tering

-0.70

design

-0.69

atin

-0.69

ergy

-0.69

clude

-0.68

corn

-0.68

mpeg

-0.67

inton

-0.64

POSITIVE LOGITS

 responsibility

0.96

 responsibilities

0.94

 delegated

0.87

 Responsibility

0.83

lessly

0.82

lessness

0.76

 respons

0.75

axter

0.74

 obligation

0.74

 owed

0.73

Activations Density 0.019%

statements emphasizing or discussing responsibilities

discussions about personal and social responsibilities

No Comments

No Known Activations