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

zone

-0.70

'>

-0.67

 territ

-0.66

orea

-0.65

Products

-0.64

ansas

-0.64

prim

-0.64

 Scope

-0.63

ornia

-0.62

ledged

-0.62

POSITIVE LOGITS

ĪĴ

0.80

Geh

0.73

 fung

0.70

atha

0.70

¥

0.68

Ĥİ

0.66

«ĺ

0.62

 merchants

0.61

½

0.61

 Budd

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.