INDEX

Explanations

salutations or greetings in the text

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

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

 omnibus

-0.64

ductos

-0.62

 Navarre

-0.62

 caufe

-0.62

 whofe

-0.62

hypothesis

-0.61

 adev

-0.59

 Nebel

-0.59

حق

-0.58

 becauſe

-0.58

POSITIVE LOGITS

Hey

1.18

Hey

1.07

Hi

1.05

Hej

0.96

Dear

0.90

Hello

0.87

hey

0.86

HEY

0.81

Hy

0.80

Hy

0.79

Activations Density 0.037%

salutations or greetings in the text

No Comments

No Known Activations