INDEX

Explanations

references to medical processes and patient support in healthcare contexts

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

 rencontre

-0.60

 poches

-0.56

ItemBackground

-0.55

Presence

-0.52

permum

-0.52

SpringBoot

-0.47

 vœux

-0.47

 hábito

-0.46

 entrées

-0.46

 ospiti

-0.45

POSITIVE LOGITS

CppMethod

0.77

มาะ

0.68

ppins

0.65

 صوتيه

0.65

AutoField

0.64

 تانيه

0.60

ChildScrollView

0.60

Witam

0.60

andExpect

0.60

 تضيفلها

0.60

Activations Density 0.124%

references to medical processes and patient support in healthcare contexts

No Comments

No Known Activations