INDEX

Explanations

phrases related to something being added to an existing situation or entity

oai_token-act-pair · gpt-3.5-turbo

phrases indicating something that is already established or well-known

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

eday

-0.88

 Females

-0.78

ĺħ

-0.77

 Tags

-0.76

oons

-0.71

 Dreams

-0.70

ODY

-0.69

 Shutterstock

-0.69

 Lines

-0.68

 RELEASE

-0.67

POSITIVE LOGITS

 tumultuous

1.04

 chaotic

1.03

 shaky

0.98

 turbulent

0.98

 formidable

0.97

 insignificant

0.97

 illustrious

0.97

 meager

0.97

 lucrative

0.96

 volatile

0.96

Activations Density 0.141%

phrases related to something being added to an existing situation or entity

phrases indicating something that is already established or well-known

No Comments

No Known Activations