INDEX

Explanations

No Explanations Found

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rences

-0.93

uary

-0.76

thens

-0.75

iability

-0.73

erb

-0.72

Â¶

-0.71

lees

-0.71

ston

-0.70

ooks

-0.69

DH

-0.69

POSITIVE LOGITS

 Predator

0.67

 Taxi

0.66

 advoc

0.66

 Marketable

0.62

 Lover

0.62

 Truck

0.61

 welf

0.60

 Belief

0.60

 Fundamental

0.58

 vain

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.