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

RF

-0.69

iga

-0.68

alogy

-0.67

 Facilities

-0.64

Vs

-0.64

 Houses

-0.63

 franchise

-0.63

Store

-0.62

istan

-0.61

HL

-0.61

POSITIVE LOGITS

ntil

0.81

ategory

0.74

worn

0.73

 unlaw

0.71

ailability

0.68

armed

0.68

hedon

0.67

rost

0.67

Rod

0.67

pter

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.