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

CHAT

-0.69

"]=>

-0.68

Ñı

-0.67

ria

-0.67

TPS

-0.66

×Ļ×

-0.65

XY

-0.65

ãĥ¢

-0.64

ilty

-0.63

ppo

-0.63

POSITIVE LOGITS

itent

0.76

henko

0.73

 administ

0.72

hyde

0.69

iferation

0.67

claimer

0.64

hops

0.64

lyr

0.64

 diplomat

0.63

 smugg

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.