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

CT

-0.72

acea

-0.72

assin

-0.70

tein

-0.69

tz

-0.69

 Citiz

-0.67

[/

-0.65

liest

-0.64

hens

-0.64

atics

-0.63

POSITIVE LOGITS

 barriers

0.66

depend

0.65

 ceilings

0.62

products

0.62

 Rite

0.61

Dir

0.60

 classmate

0.59

 Chambers

0.58

ooks

0.58

 Mile

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.