INDEX

Explanations

phrases related to actions and their consequences

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

down on

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

bluza

-0.52

nourriture

-0.51

 Membre

-0.49

affaires

-0.47

hdashline

-0.46

 grecque

-0.46

 gonzález

-0.46

mochila

-0.46

 bénévol

-0.45

JSONObject

-0.45

POSITIVE LOGITS

RegistryLite

0.63

 nahilalakip

0.59

 upon

0.52

 ***!

0.51

GW

0.51

writeFieldEnd

0.51

 emphasis

0.49

 reliance

0.47

 UPON

0.46

 relied

0.46

Activations Density 0.040%

phrases related to actions and their consequences

down on

No Comments

No Known Activations

phrases related to actions and their consequences

down on

No Comments

No Known Activations