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

alone

-0.69

hom

-0.68

dating

-0.68

XII

-0.67

heter

-0.66

plet

-0.66

dom

-0.65

ocl

-0.63

 fingert

-0.63

 XIII

-0.63

POSITIVE LOGITS

 Annotations

0.70

 Anxiety

0.65

arton

0.63

oller

0.63

 externalToEVAOnly

0.61

[_

0.60

ayer

0.60

Provider

0.60

 adaptive

0.59

SourceFile

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.