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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

regon

-0.78

sie

-0.69

advertisement

-0.67

iamond

-0.62

amin

-0.62

¬¼

-0.61

 Nicole

-0.60

 witchcraft

-0.58

hler

-0.58

querque

-0.58

POSITIVE LOGITS

 burd

0.74

FUL

0.72

...]

0.70

terday

0.69

fare

0.67

 bearer

0.66

theless

0.65

ANK

0.65

ockets

0.64

 COUN

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.