INDEX

Explanations

details about the length and structure of films or documentaries

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

NextToken

-0.43

bezpe

-0.40

harapkan

-0.36

Viited

-0.35

 particuliers

-0.34

 Italij

-0.34

escaped

-0.33

antaranya

-0.33

OBJETIVOS

-0.33

tigkeiten

-0.32

POSITIVE LOGITS

 noDo

0.72

 فريبيس

0.65

twimg

0.65

 Roskov

0.59

UrlResolution

0.58

figsize

0.58

ConstraintMaker

0.57

 Signalez

0.57

AutoField

0.56

WebElementEntity

0.56

Activations Density 0.012%

details about the length and structure of films or documentaries

No Comments

No Known Activations