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

inge

-0.68

 lame

-0.67

 reck

-0.67

market

-0.64

 dismissive

-0.64

gery

-0.64

oday

-0.61

gotten

-0.60

idy

-0.58

 hacked

-0.58

POSITIVE LOGITS

ãĤ¨ãĥ«

0.84

ModLoader

0.84

ãĥ¯ãĥ³

0.81

TPPStreamerBot

0.80

åŃĲ

0.79

å§«

0.79

éĹĺ

0.77

bilt

0.77

ä¹

0.77

è»

0.77

Activations Density 0.000%

No Known Activations

This feature has no known activations.