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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uana

-0.76

istor

-0.73

itory

-0.73

oice

-0.72

achev

-0.70

anguage

-0.70

imir

-0.69

etsk

-0.67

otte

-0.67

jong

-0.67

POSITIVE LOGITS

0.78

 Precision

0.69

 Celt

0.67

CLR

0.67

Temperature

0.65

whelming

0.64

Done

0.64

 Nokia

0.63

Þ

0.63

 Ended

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.