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

tops

-0.89

gam

-0.79

pai

-0.75

istics

-0.72

warm

-0.70

fell

-0.70

zees

-0.68

ricular

-0.68

xit

-0.66

cand

-0.66

POSITIVE LOGITS

 variance

0.66

 Synd

0.65

arcity

0.65

 skelet

0.64

abulary

0.62

 unbeliev

0.61

 veter

0.60

emetery

0.60

 compr

0.58

 Collider

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.