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

glers

-0.73

gered

-0.69

 Surprise

-0.67

ãģį

-0.67

 Rasm

-0.65

 Volks

-0.64

uple

-0.64

 iceberg

-0.64

 Bild

-0.63

Boe

-0.63

POSITIVE LOGITS

hire

0.65

ksh

0.62

iah

0.61

ension

0.61

aton

0.60

isites

0.60

letters

0.60

 rehabilit

0.60

hend

0.59

iew

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.