INDEX

Explanations

expressions related to procedural instructions or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/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.31.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

GOTREF

-0.50

Rüyada

-0.39

 ligiloj

-0.33

nu

-0.32

 surla

-0.31

本

-0.31

Ce

-0.31

multicolumn

-0.31

Ce

-0.31

em

-0.31

POSITIVE LOGITS

 isso

1.81

これは

1.61

 это

1.54

 cela

1.42

这一点

1.41

 esto

1.40

 ذلك

1.39

 Isso

1.31

这点

1.30

 ello

1.27

Activations Density 2.124%

expressions related to procedural instructions or processes

No Comments

No Known Activations