INDEX

Explanations

occurrences of non-verbal expressions of certainty or importance

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

position, column, CHECK, pragma

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

ngrx

-0.56

 ProtoMessage

-0.53

 MutableLiveData

-0.53

@"/

-0.50

----</

-0.50

новниш

-0.49

 pytest

-0.49

 OnDestroy

-0.48

 iParam

-0.47

conscious

-0.47

POSITIVE LOGITS

ParallelGroup

0.48

 introducing

0.39

 introduction

0.37

 construction

0.36

 introduced

0.35

 department

0.35

 introd

0.35

 introduces

0.34

Introducing

0.34

 Introducing

0.34

Activations Density 0.001%

occurrences of non-verbal expressions of certainty or importance

position, column, CHECK, pragma

No Comments

No Known Activations