INDEX

Explanations

references to mathematical or numerical data and models

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ब्रेकडाउन

-0.52

 Roskov

-0.52

twimg

-0.51

 المعيارى

-0.49

 hóa

-0.49

 stockbild

-0.46

 queſta

-0.45

FunctionFlags

-0.45

XmlAccessType

-0.44

tocin

-0.44

POSITIVE LOGITS

 Ingeniería

0.62

 engineering

0.59

 Engineering

0.58

 engineers

0.56

 ingenieros

0.56

Engineering

0.53

 Ingenie

0.53

 ingeniería

0.53

engineering

0.52

 engineer

0.51

Activations Density 0.510%

references to mathematical or numerical data and models

No Comments

No Known Activations