INDEX

Explanations

email addresses and mentions in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 reacties

-0.42

 moeite

-0.41

 reaſon

-0.40

もしれません

-0.39

ſelf

-0.39

telar

-0.39

 NDEBUG

-0.39

ſelves

-0.38

 Majefty

-0.38

 sanitarios

-0.38

POSITIVE LOGITS

 AssemblyCompany

0.83

://

0.78

FieldBuilder

0.59

0.56

 getItemId

0.54

 فريبيس

0.52

 resourceCulture

0.51

ษัท

0.51

MessageOf

0.50

Activations Density 0.181%

email addresses and mentions in the text

No Comments

No Known Activations