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

 Fiorina

-0.72

 challeng

-0.69

rose

-0.67

nah

-0.66

 Antar

-0.65

 Fitness

-0.64

 Mahm

-0.63

 Hayward

-0.63

 Robbins

-0.63

deen

-0.62

POSITIVE LOGITS

 ignition

0.74

hement

0.73

angan

0.69

idon

0.69

Greek

0.66

irl

0.65

 Hung

0.64

imity

0.63

POR

0.62

 underway

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.