INDEX

Explanations

repeated mentions or calls for action

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

calls for action or entities

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

tanleria

-0.49

 Савезне

-0.45

jenige

-0.43

 initComponents

-0.42

ymce

-0.42

 réussir

-0.41

 Hälfte

-0.41

 potato

-0.41

 Mahesh

-0.40

 isComment

-0.40

POSITIVE LOGITS

call

0.66

CALL

0.62

 call

0.60

 Call

0.60

 CALL

0.59

Call

0.59

 Calling

0.50

 calling

0.49

Calling

0.48

calling

0.48

Activations Density 0.006%

repeated mentions or calls for action

calls for action or entities

No Comments

No Known Activations