INDEX

Explanations

instances of the word "compute" and its various forms indicating computation or processing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 MainAxisSize

-0.88

 Adela

-0.83

 Havel

-0.80

Illuminate

-0.73

 Rasul

-0.71

 Schuster

-0.70

makeConstraints

-0.69

 Haga

-0.69

Укра

-0.68

 Rapp

-0.68

POSITIVE LOGITS

queryInterface

0.86

 σε

0.85

von

0.83

 Pollack

0.82

Von

0.77

anyika

0.77

ruptcy

0.73

icidio

0.73

]));

0.72

 Redmond

0.72

Activations Density 0.045%

instances of the word "compute" and its various forms indicating computation or processing actions

No Comments

No Known Activations