INDEX

Explanations

references to evaluation methods and effectiveness in research studies

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

comparing methods and treatments

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

protoimpl

-0.54

 houſe

-0.41

 semmi

-0.38

GOTREF

-0.36

setVerticalGroup

-0.35

 Diſ

-0.35

 oprot

-0.35

 purpoſe

-0.34

MemoryWarning

-0.33

scriptcase

-0.33

POSITIVE LOGITS

 alternative

0.56

enido

0.52

 methods

0.49

ⓧ

0.48

Hozzáférés

0.48

WriteAttribute

0.48

 treatments

0.47

-------

0.46

<?

0.45

 alternatives

0.45

Activations Density 0.109%

references to evaluation methods and effectiveness in research studies

comparing methods and treatments

No Comments

No Known Activations