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

duration

-0.78

ceed

-0.70

Äĩ

-0.69

 Edit

-0.68

upt

-0.67

uber

-0.63

ging

-0.63

inct

-0.62

igure

-0.62

uge

-0.62

POSITIVE LOGITS

untled

0.64

ilet

0.61

idental

0.61

 veiled

0.58

 WRITE

0.57

orah

0.57

ahi

0.55

 confession

0.55

cffffcc

0.55

 Coul

0.54

Activations Density 0.000%

No Known Activations

This feature has no known activations.