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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Earn

-0.66

ãĥĪ

-0.65

atts

-0.64

adeon

-0.63

ÑĤ

-0.62

 Earn

-0.62

veland

-0.61

ãĥ©

-0.60

iltr

-0.60

ãĥ«

-0.59

POSITIVE LOGITS

 above

0.85

udic

0.81

ĸļ

0.77

lest

0.70

ortment

0.66

bul

0.65

paralle

0.65

improve

0.64

 cous

0.64

above

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.