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

ead

-0.79

ounding

-0.71

uden

-0.70

lyak

-0.69

umn

-0.68

aukee

-0.68

itas

-0.68

chat

-0.67

otic

-0.67

idine

-0.67

POSITIVE LOGITS

 coerc

0.79

 withd

0.76

allery

0.73

 surrounds

0.68

 accompanies

0.66

 cumbers

0.66

 contradicts

0.66

 carbohyd

0.65

 entert

0.64

 connects

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.