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

luent

-0.80

opian

-0.77

inea

-0.74

igmat

-0.73

Ĥİ

-0.73

elight

-0.73

ographs

-0.73

lear

-0.73

urnal

-0.70

inventoryQuantity

-0.69

POSITIVE LOGITS

 Marg

0.72

Moh

0.65

 Surviv

0.64

 Directorate

0.63

 Estimates

0.63

 RAND

0.63

 IPCC

0.61

 profile

0.60

Reb

0.60

 Meteor

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.