INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 cler

-0.72

 routed

-0.64

 playbook

-0.63

Ļ½

-0.63

 Commun

-0.62

 reperto

-0.61

 rÃ©

-0.60

cdn

-0.60

 tender

-0.59

 reception

-0.58

POSITIVE LOGITS

ocobo

0.77

0.75

farm

0.72

float

0.70

emp

0.68

[&

0.68

oshi

0.68

argon

0.67

oidal

0.66

dor

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.