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

rament

-0.89

ogle

-0.77

mot

-0.77

gae

-0.75

iba

-0.74

adin

-0.74

oki

-0.73

ja

-0.72

undai

-0.71

amina

-0.71

POSITIVE LOGITS

 reciproc

0.84

 Jobs

0.76

 Crack

0.72

 Collabor

0.70

INESS

0.67

ERG

0.65

 Continuous

0.64

è¿

0.64

 OPEN

0.64

WARD

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.