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

battle

-0.74

Center

-0.71

istics

-0.71

uality

-0.69

iasis

-0.69

atos

-0.69

archy

-0.68

ista

-0.67

oward

-0.66

ace

-0.65

POSITIVE LOGITS

 epis

0.69

 newsp

0.65

rogens

0.63

 showc

0.63

 parole

0.63

 accordingly

0.62

 traged

0.60

VERTIS

0.60

 wiret

0.60

inventoryQuantity

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.