INDEX

Explanations

descriptions related to skill or talent

oai_token-act-pair · gpt-3.5-turbo

references to various skills and abilities in a performance context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Reloaded

-0.75

 EDITION

-0.66

 Corinthians

-0.65

 Amen

-0.65

 Manz

-0.63

 Express

-0.63

 Sagan

-0.62

 Cosby

-0.61

ODUCT

-0.61

Wem

-0.60

POSITIVE LOGITS

fully

1.13

anooga

1.12

ful

0.98

poke

0.92

 skills

0.85

smanship

0.85

chery

0.84

reys

0.80

 skill

0.79

cius

0.77

Activations Density 0.017%

descriptions related to skill or talent

references to various skills and abilities in a performance context

No Comments

No Known Activations