INDEX

Explanations

references to written content like articles or statements

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 closures

-0.70

 groundwork

-0.70

 roadmap

-0.67

 demos

-0.67

Thumbnail

-0.63

 moons

-0.63

 timelines

-0.63

 basics

-0.63

 albums

-0.62

 essentials

-0.62

POSITIVE LOGITS

his

0.97

him

0.94

he

0.92

He

0.91

He

0.87

 Another

0.85

 respondent

0.82

His

0.81

 Someone

0.81

His

0.80

Activations Density 0.488%

references to written content like articles or statements

phrases indicating significant events or actions

No Comments

No Known Activations