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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Seym

-0.75

 mathemat

-0.68

 handshake

-0.67

Split

-0.66

 wors

-0.66

 poppy

-0.65

 portrayal

-0.64

 scapego

-0.63

 bond

-0.62

 drib

-0.61

POSITIVE LOGITS

abeth

0.72

ographers

0.71

 Whitman

0.69

ographer

0.69

tml

0.66

vana

0.65

 leukemia

0.65

^^^^

0.63

aspx

0.63

^^

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.