INDEX

Explanations

references to judicial reports and recommendations

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

report and recommendation

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

RegressionTest

-0.42

rungsseite

-0.37

Begriffsklä

-0.35

PostBack

-0.34

 symp

-0.34

Przypisy

-0.34

IntoConstraints

-0.34

 utafitiHapana

-0.33

lucky

-0.33

 resourceCulture

-0.33

POSITIVE LOGITS

 recommendation

0.60

styleType

0.58

 recommendations

0.55

recommendation

0.55

 recommends

0.55

 recomend

0.54

tagHelperRunner

0.54

recommendations

0.54

RECOMMEND

0.53

 recommending

0.53

Activations Density 0.041%

references to judicial reports and recommendations

report and recommendation

No Comments

No Known Activations