INDEX

Explanations

technical terms and elements related to coding or data analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SourceChecksum

-0.65

 initComponents

-0.55

styleType

-0.53

Hentet

-0.49

ंदीखरीदारी

-0.48

 ProtoMessage

-0.46

ruvate

-0.45

 vaan

-0.45

 ModelExpression

-0.45

elemField

-0.44

POSITIVE LOGITS

 ſeveral

0.45

 Reſ

0.43

 AspNetCore

0.42

ESSOR

0.42

 viſ

0.41

 ISNI

0.41

 tranſ

0.40

 verſ

0.40

CROSSTALK

0.39

 themſelves

0.39

Activations Density 0.408%

technical terms and elements related to coding or data analysis

No Comments

No Known Activations