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

eday

-0.98

govtrack

-0.80

 Scrib

-0.71

 Nieto

-0.71

nir

-0.67

 unification

-0.67

iosyncr

-0.66

stru

-0.66

 sovereignty

-0.64

bom

-0.64

POSITIVE LOGITS

glers

0.79

 props

0.66

Î

0.62

pin

0.62

cised

0.61

Cass

0.61

iths

0.60

picked

0.59

mates

0.58

Î¿

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.