INDEX

Explanations

references to rankings, records, and achievements in various contexts, particularly in gaming and sports

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

webElementXpaths

-0.54

Rhestr

-0.50

 vPvB

-0.49

Izvori

-0.48

šť

-0.48

óch

-0.47

KURZBESCHREIBUNG

-0.47

šn

-0.46

 estekak

-0.45

 nahilalakip

-0.45

POSITIVE LOGITS

all

1.38

all

1.03

 ever

0.95

ALL

0.84

ever

0.78

All

0.78

ALL

0.75

All

0.75

 lifetime

0.75

 EVER

0.74

Activations Density 0.099%

references to rankings, records, and achievements in various contexts, particularly in gaming and sports

No Comments

No Known Activations