INDEX

Explanations

expressions related to surprise or unexpectedness

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

Following words indicating surprise or noteworthiness

np_acts-logits-general · gemini-2.0-flash

surprising or notable observations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

IUrlHelper

-0.80

esModule

-0.66

StoryboardSegue

-0.65

addCriterion

-0.61

 noqa

-0.58

TestTools

-0.57

onViewCreated

-0.56

MigrationBuilder

-0.55

曖昧さ回避

-0.55

 ControllerBase

-0.54

POSITIVE LOGITS

 considering

1.46

considering

1.27

Considering

1.24

 Considering

1.23

 given

1.18

 mengingat

1.14

 учиты

1.06

 considerando

0.99

Given

0.98

考虑到

0.96

Activations Density 0.373%

expressions related to surprise or unexpectedness

Following words indicating surprise or noteworthiness

surprising or notable observations

No Comments

No Known Activations