INDEX

Explanations

phrases indicating reference or connection to related topics

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

links and categories

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

<eos>

-0.39

Bru

-0.38

 Cooper

-0.38

Encoded

-0.38

-0.37

 Bruch

-0.36

 propio

-0.36

 secretly

-0.35

POSITIVE LOGITS

 disambiguazione

0.83

 autorytatywna

0.81



0.76

verwijspagina

0.70

AddTagHelper

0.69

ביוגרפיה

0.65

rungsseite

0.64

曖昧さ回避

0.62

PreferredItem

0.59

RTLR

0.59

Activations Density 0.006%

phrases indicating reference or connection to related topics

links and categories

No Comments

No Known Activations