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

Vari

-0.73

Premium

-0.72

Fav

-0.69

 Amend

-0.69

 Tactics

-0.67

[_

-0.66

 Marriage

-0.66

ãĥ¼ãĥ³

-0.66

Dial

-0.66

Kings

-0.64

POSITIVE LOGITS

 cooler

0.77

 mafia

0.71

 booted

0.67

 logged

0.63

 firefighter

0.63

 warn

0.62

 bullies

0.62

 torch

0.62

 weep

0.62

 logging

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.