INDEX

Explanations

references to studies, cases, and specific identifiers related to research and legal documentation

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

research papers and systems

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



-0.81

 nakalista

-0.74

 مشين

-0.71

GOTREF

-0.68

Hentet

-0.66

 propOrder

-0.59

 nonUne

-0.58

 ObjectParameter

-0.57

 المعيارى

-0.55

RegressionTest

-0.55

POSITIVE LOGITS

 <>",

0.38

ebenarnya

0.38

 Constitucional

0.37

 dedicar

0.36

เอง

0.36

 Decke

0.36

 kese

0.35

 Mischung

0.35

 coser

0.35

 Dichtung

0.35

Activations Density 0.027%

references to studies, cases, and specific identifiers related to research and legal documentation

research papers and systems

No Comments

No Known Activations