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

awa

-0.89

iator

-0.76

¥µ

-0.72

iable

-0.71

 Dimension

-0.64

 advant

-0.64

idable

-0.63

 reliable

-0.63

 predictor

-0.62

iple

-0.61

POSITIVE LOGITS

cheon

0.78

icum

0.77

 actionGroup

0.74

acter

0.70

 Gingrich

0.70

quit

0.69

elected

0.67

PAC

0.64

Ops

0.64

mic

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.