INDEX

Explanations

references to multiple languages and their translations or adaptations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

tagHelperRunner

-0.62

 الرياضيه

-0.59

 Roskov

-0.57

intios

-0.56

TagHelper

-0.53

تفصیلات

-0.52

RenderAtEndOf

-0.52

Хьажоргаш

-0.50

DockStyle

-0.50

StoryboardSegue

-0.50

POSITIVE LOGITS

 estekak

0.39

エスト

0.34

 Jakarta

0.34

ABCD

0.32

 erfü

0.32

dymo

0.32

وعة

0.31

テム

0.31

English

0.31

ſelf

0.30

Activations Density 0.013%

references to multiple languages and their translations or adaptations

No Comments

No Known Activations