INDEX

Explanations

expressions of location and reference to specific places

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

affirmation or agreement

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

 defaultstate

-0.61

 desmotivaciones

-0.60

 Pardavimas

-0.56

 MessageBoxIcon

-0.54

 snippetHide

-0.53

Tikang

-0.52

__":

-0.51

 ſche

-0.51

 indígen

-0.50

addContainerGap

-0.49

POSITIVE LOGITS

 nahilalakip

0.41

ox

0.39

 Uptake

0.36

 bParam

0.36

 Normdatei

0.34

<!--[

0.34

 ostavi

0.33

ParallelGroup

0.33

Šaltiniai

0.33

co

0.33

Activations Density 0.011%

expressions of location and reference to specific places

affirmation or agreement

No Comments

No Known Activations