INDEX

Explanations

specific numerical and categorical data points related to health, behaviors, and statistical metrics in a dataset

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

[{

-0.48

жнему

-0.47

WebVitals

-0.47

 tambin

-0.46

 extranjeros

-0.46

RegistryLite

-0.46

 initComponents

-0.45

 permukaan

-0.44

 keskus

-0.44

 définiti

-0.43

POSITIVE LOGITS

󠁴

0.47

RTEX

0.43

cables

0.38

DoubleQuotes

0.38

 ModelRenderer

0.38

OGND

0.36

 Signalez

0.36

Aholisi

0.36

titleMargin

0.36

muer

0.36

Activations Density 0.084%

specific numerical and categorical data points related to health, behaviors, and statistical metrics in a dataset

No Comments

No Known Activations