INDEX

Explanations

references to comparative data and relationships

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

prepositions followed by nouns

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

 BoxFit

-0.68

лтамалар

-0.59

RTGC

-0.54

MST

-0.54

InputTagHelper

-0.53

EDD

-0.52

HostException

-0.51

WaitGroup

-0.51

úrese

-0.50

transQ

-0.50

POSITIVE LOGITS

 others

0.40

 stuks

0.38

 demais

0.36

 právě

0.35

 keduanya

0.34

 doméstico

0.32

 actuelles

0.32

 även

0.32

 meeste

0.32

 umumnya

0.31

Activations Density 0.071%

references to comparative data and relationships

prepositions followed by nouns

No Comments

No Known Activations