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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hene

-0.67

HCR

-0.66

served

-0.65

apr

-0.65

adden

-0.64

ocious

-0.64

Higher

-0.64

stasy

-0.61

STEM

-0.61

eva

-0.61

POSITIVE LOGITS

 pause

0.67

################

0.66

iser

0.66

 Seller

0.65

 inquire

0.64

 attribution

0.63

lus

0.62

estate

0.62

river

0.60

 Loyal

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.