INDEX

Explanations

structure and formatting elements in a document

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

 ModelExpression

-0.50

 dezelve

-0.45

WriteTagHelper

-0.41

 pouvoit

-0.40

 tombs

-0.40

entraînement

-0.40

 Links

-0.39

Égypte

-0.38

égories

-0.38

Québec

-0.36

POSITIVE LOGITS

 Roskov

0.53

 disambiguazione

0.52

mybatisplus

0.52

 Савезне

0.50

MessageOf

0.49

 المعيارى

0.48

 للمعارف

0.48

abstractmethod

0.48

tvguidetime

0.47

 propOrder

0.46

Activations Density 0.519%

structure and formatting elements in a document

No Comments

No Known Activations