INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ozyg

-0.79

aird

-0.73

 Instr

-0.67

jl

-0.67

ograp

-0.66

 Hound

-0.66

isp

-0.64

 Scand

-0.63

efficiency

-0.61

 Widow

-0.60

POSITIVE LOGITS

soever

0.80

BuyableInstoreAndOnline

0.79

ILY

0.78

~~~~

0.71

avage

0.71

rawdownloadcloneembedreportprint

0.71

âĶĢâĶĢâĶĢâĶĢ

0.69

VERTISEMENT

0.68

FORMATION

0.66

terday

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.