INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

usalem

-0.75

haw

-0.70

 Ambassador

-0.69

 McConnell

-0.69

 Ministers

-0.68

 Seat

-0.67

 Sabha

-0.64

 Pilgrim

-0.64

BJP

-0.63

Maw

-0.61

POSITIVE LOGITS

enne

0.98

erm

0.94

phia

0.81

ences

0.80

teness

0.79

erity

0.79

racted

0.79

ensing

0.78

yne

0.74

arc

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.