INDEX

Explanations

qualitative assessments of experiences or items, particularly in a review context

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

positive sentiment adjectives

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

enterOuterAlt

-0.44

исленность

-0.43

RefNanny

-0.40

ScopeManager

-0.39

jsii

-0.38

resave

-0.38

webElementXpaths

-0.36

 Manufacturing

-0.36

ocrats

-0.35

 Approach

-0.35

POSITIVE LOGITS

 otomatig

0.57

oredCriteria

0.46

 QLabel

0.45

httphttps

0.45

GeoNames

0.44

壓

0.43

 âmes

0.43

exitRule

0.43

 increí

0.43

 autorytatywna

0.42

Activations Density 0.025%

qualitative assessments of experiences or items, particularly in a review context

positive sentiment adjectives

No Comments

No Known Activations