INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

imilation

-0.69

oteric

-0.65

 Tray

-0.65

 consciously

-0.63

 Indigenous

-0.63

phy

-0.62

ective

-0.62

itic

-0.61

elong

-0.60

atron

-0.60

POSITIVE LOGITS

esta

0.75

esc

0.74

assets

0.67

EVA

0.65

launch

0.63

gins

0.61

wal

0.61

hig

0.61

ewater

0.60

etus

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.