INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ortium

-0.76

ospace

-0.71

aditional

-0.69

 astroph

-0.69

urgical

-0.69

 surrogate

-0.68

urgy

-0.68

abwe

-0.67

ighters

-0.64

urat

-0.63

POSITIVE LOGITS

LET

0.68

Fas

0.67

asing

0.66

|--

0.65

+#

0.65

 attRot

0.63

###

0.62

Site

0.61

OS

0.60

[+

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.