INDEX

Explanations

words related to specific skills or abilities

oai_token-act-pair · gpt-3.5-turbo

references to skill levels and abilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Wem

-0.70

 Amen

-0.70

 Corinthians

-0.69

 Pall

-0.67

 Manz

-0.65

Jane

-0.65

 Cosby

-0.65

 Reloaded

-0.64

 Plymouth

-0.63

 Dover

-0.62

POSITIVE LOGITS

anooga

1.01

fully

0.95

poke

0.95

ful

0.91

smanship

0.86

 guiActiveUn

0.84

 skills

0.82

Skill

0.81

 skill

0.80

uracy

0.79

Activations Density 0.011%

words related to specific skills or abilities

references to skill levels and abilities

No Comments

No Known Activations