INDEX

Explanations

numerical data and statistical information

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

脚注の使い方

-0.95

StoreMessageInfo

-0.75

########.

-0.70

IUrlHelper

-0.70

 defaultstate

-0.68

WriteTagHelper

-0.67

 oprot

-0.67

 kasarigan

-0.67

 OFDb

-0.67

setVerticalGroup

-0.64

POSITIVE LOGITS

 burton

0.43

SSM

0.41

šče

0.40

+"\

0.39

Carg

0.39

pung

0.39

sdb

0.38

 maal

0.38

 occidentalis

0.38

Burton

0.37

Activations Density 0.439%

numerical data and statistical information

No Comments

No Known Activations