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

IDs

-0.77

TM

-0.74

ICAN

-0.73

Introduced

-0.72

GOP

-0.70

OT

-0.69

Drag

-0.64

OTE

-0.62

ALS

-0.62

LV

-0.62

POSITIVE LOGITS

auga

0.78

teness

0.73

inately

0.70

 Shroud

0.68

yip

0.68

 brill

0.68

 Bung

0.67

 bureau

0.66

 Leilan

0.65

ģĸ

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.