INDEX

Explanations

specific character names from stories or films

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

people's names

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

KommentareTeilen

-0.44

cussion

-0.40

Relevant

-0.40

makeConstraints

-0.40

躇

-0.39

 AssemblyCulture

-0.39

lusconi

-0.39

 Explicit

-0.38

isha

-0.38

IntoConstraints

-0.38

POSITIVE LOGITS

+#+#

0.69

 OMITBAD

0.61

tanleria

0.57

 otomatig

0.54

LLocation

0.51

 sumpay

0.51

ագրություններ

0.50

 ItemStack

0.49

 RoutedEventArgs

0.48

OGND

0.48

Activations Density 0.034%

specific character names from stories or films

people's names

No Comments

No Known Activations