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

ioch

-0.80

 judiciary

-0.75

 mosqu

-0.74

alore

-0.73

awaru

-0.70

initialized

-0.68

 unpop

-0.68

 circumference

-0.66

 courthouse

-0.63

rero

-0.63

POSITIVE LOGITS

DER

0.85

idious

0.82

 Tycoon

0.80

zhen

0.71

ãĤ¦ãĤ¹

0.70

orney

0.69

ãĥ£

0.69

ency

0.68

ãĤ´ãĥ³

0.68

.<

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.