INDEX

Explanations

expressions of personal involvement or opinions

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

start of user input

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

windowFixed

-0.43

bootstrapcdn

-0.43

 saites

-0.42

 doInBackground

-0.41

 <>",

-0.41

 >=",

-0.41

 tartalomajánló

-0.41

MethodManager

-0.40

 tenisky

-0.39

hicles

-0.38

POSITIVE LOGITS

OGND

0.53



0.50

 Sorgen

0.45

 disruptive

0.44

Дереккөздер

0.44

SharedDtor

0.43

True

0.42

tabular

0.42

Urs

0.42

///</

0.42

Activations Density 0.006%

expressions of personal involvement or opinions

start of user input

No Comments

No Known Activations