INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

nings

-0.81

gas

-0.81

cest

-0.76

oller

-0.74

mur

-0.73

ress

-0.70

ning

-0.69

azz

-0.68

anoia

-0.68

aths

-0.68

POSITIVE LOGITS

 newfound

0.74

 particular

0.72

 antioxid

0.70

 matters

0.69

 Matters

0.67

 latter

0.66

 constituted

0.65

 amounts

0.62

 millenn

0.62

 article

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.