INDEX

Explanations

words related to skills or abilities

oai_token-act-pair · gpt-3.5-turbo

references to skills and skill levels

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

 Reloaded

-0.76

 Manz

-0.75

 Corinthians

-0.74

 Amen

-0.71

 EDITION

-0.70

romeda

-0.69

clipse

-0.68

 Cosby

-0.63

Wem

-0.61

 Sagan

-0.61

POSITIVE LOGITS

fully

1.06

anooga

1.01

ful

0.98

poke

0.96

learn

0.90

cius

0.87

chery

0.83

fighter

0.83

smanship

0.81

 skills

0.81

Activations Density 0.023%

words related to skills or abilities

references to skills and skill levels

No Comments

No Known Activations