INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

counter

-0.63

 worrying

-0.63

arget

-0.62

Merit

-0.62

 promising

-0.61

nes

-0.61

 evid

-0.60

 smir

-0.60

 constitu

-0.59

 thanking

-0.59

POSITIVE LOGITS

 eleph

0.87

urat

0.77

 Seym

0.75

imester

0.74

 Tune

0.73

 Elephant

0.73

 Dolphin

0.69

rontal

0.68

ô

0.68

undai

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.