INDEX

Explanations

descriptive elements related to characters and their appearances in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 informée

-0.79

WireFormatLite

-0.78

 nakalista

-0.75

脚注の使い方

-0.69

нгред

-0.67



-0.67

 transfieras

-0.66

﹧

-0.66

 transférez

-0.65

 zwiſchen

-0.65

POSITIVE LOGITS

 ${({

0.31

 onOptions

0.27

 poor

0.27

 ferons

0.26

zve

0.26

BCS

0.25

OptionsMenu

0.24

vyu

0.24

 figliu

0.23

 clip

0.23

Activations Density 0.198%

descriptive elements related to characters and their appearances in narratives

No Comments

No Known Activations