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

netflix

-0.73

hillary

-0.72

Qaida

-0.72

dayName

-0.71

Assembly

-0.70

Sample

-0.69

ready

-0.67

detail

-0.67

enough

-0.67

Methods

-0.66

POSITIVE LOGITS

ochond

0.67

alus

0.64

 associates

0.64

 engulf

0.64

ammers

0.64

 dictate

0.62

LLOW

0.61

 associate

0.60

ensable

0.60

 separat

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.