INDEX

Explanations

phrases related to online services and support, particularly for drivers

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

 Braun

-0.42

 typelib

-0.42

 strangled

-0.42

 Pueden

-0.42

 türlü

-0.42

 voran

-0.41

abas

-0.41

 undocumented

-0.41

iformis

-0.39

 conceded

-0.39

POSITIVE LOGITS

 kasarigan

0.65

Personendaten

0.54

 informée

0.53

 surla

0.49

 Roskov

0.48

 Autorisations

0.48

 الحره

0.46

 ſche

0.45

 Reverso

0.42

CloseOperation

0.42

Activations Density 0.013%

phrases related to online services and support, particularly for drivers

No Comments

No Known Activations