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

PART

-0.67

 Reid

-0.66

 partName

-0.64

 Regulations

-0.63

 Tactics

-0.60

Topic

-0.59

 enrichment

-0.58

-0.57

inous

-0.57

 promulg

-0.57

POSITIVE LOGITS

oldown

0.82

qus

0.77

ppo

0.75

ammy

0.74

levard

0.69

ikawa

0.69

arta

0.68

ught

0.67

ased

0.66

hurst

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.