INDEX

Explanations

references to speech-related concepts and activities

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 зла

-0.39

 lowered

-0.32

 skeptical

-0.31

GrantedAuthority

-0.30

 wondering

-0.28

 darkness

-0.28

blumen

-0.27

हरू

-0.26

 wondered

-0.26

 doubted

-0.26

POSITIVE LOGITS

RenderAtEndOf

0.75

ंदीखरीदारी

0.72

 disambiguazione

0.72

<pad>

0.71

<unused42>

0.71

<unused16>

0.70

<unused79>

0.70

[@BOS@]

0.70

<unused3>

0.70

<unused14>

0.70

Activations Density 0.061%

references to speech-related concepts and activities

No Comments

No Known Activations