INDEX

Explanations

dialogue or conversational phrases, particularly involving personal interactions and responses

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

Follows a quote or speech

np_acts-logits-general · gemini-2.0-flash

reported speech or statements

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

Rohy

-0.53

StoryboardSegue

-0.51

endregion

-0.48

 zvuky

-0.46

 Bioaccumulative

-0.45

FileDescriptor

-0.43

Xar

-0.43

Cubit

-0.43

انجليز

-0.42

DeleteBehavior

-0.42

POSITIVE LOGITS

 saying

1.44

 said

1.23

say

1.16

saying

1.14

 says

1.12

Said

1.11

said

1.06

 sayin

1.05

 berkata

1.01

 Saying

1.00

Activations Density 0.160%

dialogue or conversational phrases, particularly involving personal interactions and responses

Follows a quote or speech

reported speech or statements

No Comments

No Known Activations