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

osal

-0.80

abouts

-0.66

 pores

-0.66

 bows

-0.65

 coron

-0.64

ple

-0.63

 polar

-0.60

 throats

-0.60

igent

-0.58

 confines

-0.57

POSITIVE LOGITS

 guiActiveUnfocused

0.80

 Manz

0.77

 Fallon

0.74

Ò

0.70

 Supplement

0.69

GROUP

0.67

LOAD

0.65

XP

0.64

strength

0.64

 FORM

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.