INDEX

Explanations

specific details and names in narratives or accounts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hyper

-0.46

cyd

-0.42

wenn

-0.42

 jätte

-0.41

 aufzu

-0.41

 olup

-0.41

 itself

-0.41

gebung

-0.41

 logique

-0.40

 selve

-0.40

POSITIVE LOGITS

 <=",

0.91

tagHelperRunner

0.86

ⓧ

0.83

 nahilalakip

0.79

/**

0.79

Cyfarwyddwr

0.77

 وتسجيلات

0.74

 članak

0.72

ArgsConstructor

0.70

 autorytatywna

0.70

Activations Density 0.486%

specific details and names in narratives or accounts

No Comments

No Known Activations