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

humane

-0.69

#$#$

-0.68

naissance

-0.67

stage

-0.65

event

-0.64

ighting

-0.62

portion

-0.62

 scoreboard

-0.62

ewater

-0.62

current

-0.61

POSITIVE LOGITS

abi

0.74

rely

0.72

âĹ¼

0.71

geons

0.66

Pag

0.66

 Toro

0.65

 proble

0.64

 Chand

0.63

 Sharma

0.60

illo

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.