INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

rib

-0.66

 pods

-0.65

 digs

-0.65

umbnails

-0.64

 deliveries

-0.63

raft

-0.61

omic

-0.60

ombs

-0.59

 wearer

-0.59

 whereabouts

-0.59

POSITIVE LOGITS

 Mehran

0.81

itaire

0.79

ommel

0.76

thia

0.72

ãħĭ

0.67

 Cheong

0.66

 Helpful

0.66

acebook

0.65

swick

0.65

deen

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.