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

viz

-0.71

 decline

-0.67

 spit

-0.66

 sket

-0.63

 tongues

-0.61

nib

-0.61

 Pres

-0.61

 itch

-0.61

 glory

-0.60

 Jakarta

-0.59

POSITIVE LOGITS

lio

0.84

gru

0.82

rov

0.76

aucus

0.76

 ILCS

0.76

rio

0.75

ray

0.75

itbart

0.75

 antioxid

0.75

cin

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.