INDEX

Explanations

references to sources or affiliations in a study

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

UnusedPrivate

-0.60

ſelves

-0.60

ècie

-0.59

 ſche

-0.58

RSpec

-0.58

PHE

-0.56

afficheront

-0.55

 للاسماء

-0.54

MigrationBuilder

-0.54

leſs

-0.54

POSITIVE LOGITS

the

0.57

Hochspringen

0.41

 Einfluß

0.40

مصادر

0.40

原始内容存档于

0.39

 nuvem

0.38

The

0.38

最快更新

0.38

 Kenntnis

0.37

://

0.37

Activations Density 0.025%

references to sources or affiliations in a study

No Comments

No Known Activations