INDEX

Explanations

numerical values or measurements within the text

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

scientific citations

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

TypedDataSet

-0.52

WebVitals

-0.52

addCriterion

-0.52

RTLI

-0.49

 հղումներ

-0.47

+#+

-0.46

AndroidJUnit

-0.45

 NavController

-0.45



-0.44

SharedCtor

-0.43

POSITIVE LOGITS

<sup>

0.51

<sub>

0.50

<i>

0.49

Chham

0.46

<em>

0.45

‽

0.44

<s>

0.44

<u>

0.42

swire

0.41

évaluateur

0.38

Activations Density 0.233%

numerical values or measurements within the text

scientific citations

No Comments

No Known Activations