INDEX

Explanations

references to professional titles and roles within organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Portail

-0.59

AsUp

-0.57

getDescriptor

-0.57

 StringTokenizer

-0.57

 remission

-0.56

Gön

-0.55

 незавершена

-0.55

 accruing

-0.55

ArrowToggle

-0.54

anggaran

-0.54

POSITIVE LOGITS

ResponseWriter

0.57

 Normdatei

0.57

createNewFile

0.53

 оригіналу

0.52

onded

0.52

GraphicsUnit

0.52

 resourceCulture

0.50

WebServlet

0.49

 saites

0.48

таратура

0.47

Activations Density 0.229%

references to professional titles and roles within organizations

No Comments

No Known Activations