INDEX

Explanations

scientific terms and concepts

oai_token-act-pair · gpt-3.5-turbo

references to scientific concepts and terms

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

drops

-0.75

torn

-0.70

skirts

-0.69

oning

-0.68

eu

-0.67

pora

-0.66

matched

-0.65

eper

-0.64

erous

-0.64

tower

-0.64

POSITIVE LOGITS

 fiction

1.02

 curiosity

0.97

 literacy

0.93

ĨĴ

0.89

 scientist

0.88

 research

0.86

 physicist

0.86

 breakthrough

0.86

 misconduct

0.86

 inquiry

0.85

Activations Density 0.021%

scientific terms and concepts

references to scientific concepts and terms

No Comments

No Known Activations