INDEX

Explanations

references to media and broadcasts related to specific programs and their context

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

providedIn

-0.47

 transfieras

-0.41

 Jamie

-0.41

addPreferredGap

-0.41

VersionUID

-0.41

Jamie

-0.40

AddTagHelper

-0.39

ruptedException

-0.38

ɵɵ

-0.38

 Diony

-0.38

POSITIVE LOGITS

 Monteiro

0.58

african

0.55

frica

0.55

 africano

0.55

 Uganda

0.54

 Africans

0.54

 Ugandan

0.54

African

0.52

 Liberia

0.52

 African

0.52

Activations Density 0.227%

references to media and broadcasts related to specific programs and their context

No Comments

No Known Activations