INDEX

Explanations

information related to ranking and evaluation

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

rankings or ratings

np_acts-logits-general · gemini-2.0-flash

ranking and rating

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

seitig

-0.51

 disambiguazione

-0.49

reportWebVitals

-0.48

 gyhoeddwyd

-0.47

efte

-0.46

emptyset

-0.45

 doGet

-0.45

 resourceCulture

-0.45

 dokun

-0.44

createCanvas

-0.44

POSITIVE LOGITS

 ranking

1.65

 ranked

1.65

 rank

1.60

 ranks

1.44

 rankings

1.42

rank

1.31

ranked

1.29

 Ranking

1.28

 Ranked

1.28

 Rank

1.26

Activations Density 0.263%

information related to ranking and evaluation

rankings or ratings

ranking and rating

No Comments

No Known Activations