INDEX

Explanations

sequences of numbers and special characters, likely in a bibliographic or data reference context

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

academic citations and references

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

aarrggbb

-0.43

 düz

-0.39

 dichos

-0.36

 alguna

-0.33

 ulteriori

-0.33

 Geschichten

-0.31

 Stellung

-0.31

 zéro

-0.31

 sourire

-0.31

 verdaderos

-0.30

POSITIVE LOGITS

OGND

0.79

хьтан

0.79

 transférez

0.69

Rohy

0.66

 disponibilités

0.63

Хьажоргаш

0.60

 ExecuteAsync

0.60

 EconPapers

0.59

ChildScrollView

0.57

 kasarigan

0.56

Activations Density 0.117%

sequences of numbers and special characters, likely in a bibliographic or data reference context

academic citations and references

No Comments

No Known Activations