INDEX

Explanations

phrases indicating the need to take action or make decisions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating obligation or requirements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Â®

-0.76

cus

-0.73

oshi

-0.72

WH

-0.72

OVA

-0.72

anwhile

-0.71

Synopsis

-0.71

ãĤ´

-0.70

etts

-0.68

ãĥİ

-0.66

POSITIVE LOGITS

 somebody

1.02

 gotta

0.94

 lots

0.91

 tremendous

0.91

 conversations

0.87

 tonnes

0.84

 unbelievable

0.83

 everybody

0.83

 tons

0.83

got

0.82

Activations Density 0.286%

phrases indicating the need to take action or make decisions

phrases indicating obligation or requirements

No Comments

No Known Activations