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

cheon

-0.83

hips

-0.76

etically

-0.72

iano

-0.71

anga

-0.68

inav

-0.67

ettes

-0.67

cano

-0.66

inson

-0.65

ovie

-0.64

POSITIVE LOGITS

Frames

0.65

 strand

0.64

BUG

0.63

CHAT

0.63

¥ŀ

0.62

Listener

0.62

fram

0.61

 Huck

0.60

fect

0.59

joining

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.