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

eele

-0.69

 Specialist

-0.66

 Samurai

-0.64

uffer

-0.61

osuke

-0.60

 Crush

-0.58

 specialization

-0.58

Sword

-0.57

 Colt

-0.57

XY

-0.56

POSITIVE LOGITS

iris

0.81

ebus

0.80

etheless

0.73

tle

0.73

perm

0.72

meier

0.71

tes

0.71

elight

0.70

istar

0.67

 conflic

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.