INDEX

Explanations

No Explanations Found

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

interrupted

-0.78

rawdownloadcloneembedreportprint

-0.78

 maturity

-0.77

urus

-0.77

oise

-0.73

 proble

-0.68

agher

-0.66

icum

-0.66

 wavelength

-0.64

intage

-0.63

POSITIVE LOGITS

 Brus

0.70

 Swim

0.67

erers

0.64

 Hait

0.64

opl

0.63

ings

0.63

PLA

0.63

 Toad

0.63

soever

0.61

riks

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.