INDEX

Explanations

structured data and programming constructs

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

Code snippets or bracketed expressions

np_acts-logits-general · gemini-2.0-flash

ATTRIBUTION DETAIL capability characteristic

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

Дереккөздер

-0.68

 transférez

-0.56

GREES

-0.53

 متعلقه

-0.52

Autoritní

-0.51

 rans

-0.51

 externi

-0.51

 Autorizaciones

-0.49

 autorytatywna

-0.49

ToProps

-0.49

POSITIVE LOGITS

 trait

0.83

 habit

0.81

benefit

0.71

 symptom

0.71

Symptom

0.71

 amenity

0.68

hazard

0.67

 benefit

0.67

 Symptom

0.67

 Benefit

0.66

Activations Density 0.531%

structured data and programming constructs

Code snippets or bracketed expressions

ATTRIBUTION DETAIL capability characteristic

No Comments

No Known Activations