INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

VERTISEMENT

-0.71

 elig

-0.68

METHOD

-0.67

hips

-0.65

ãĥģ

-0.64

-0.63

cens

-0.63

ESV

-0.62

rued

-0.62

ULT

-0.61

POSITIVE LOGITS

 gauge

0.74

 Lara

0.72

 Chimera

0.70

ierre

0.68

 Shroud

0.65

icter

0.64

 Fres

0.64

 Schne

0.63

imer

0.63

retty

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.