INDEX

Explanations

instances where something is waived or capped

oai_token-act-pair · gpt-3.5-turbo

terms related to legal or regulatory exemptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

oded

-0.81

ician

-0.78

elf

-0.76

ember

-0.73

 Belt

-0.72

alogue

-0.72

ophe

-0.71

icator

-0.71

aten

-0.69

 Hobby

-0.69

POSITIVE LOGITS

 waive

1.22

 waived

1.19

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

1.08

 waivers

0.75

creen

0.70

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

0.70

 warranties

0.67

ãģ®å®

0.67

 Defin

0.65

ceivable

0.65

Activations Density 0.018%

instances where something is waived or capped

terms related to legal or regulatory exemptions

No Comments

No Known Activations