INDEX

Explanations

empty or blank sections in the document

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

Code snippets containing console, class, or functions

np_acts-logits-general · gemini-2.0-flash

legal or programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

請繼續往下閱讀

-0.56

QMetaType

-0.54

fromnode

-0.52

достатки

-0.49

*/),

-0.48

ModelSerializer

-0.48

MigrationBuilder

-0.46

ReusableCell

-0.46

 otomatig

-0.46

 protoimpl

-0.45

POSITIVE LOGITS

 createSlice

0.74

<bos>

0.64

 pageContext

0.57

MCE

0.55

 Sympathy

0.54

import

0.53

mazon

0.53

entious

0.52

 Rosalie

0.52

 ministre

0.52

Activations Density 0.120%

empty or blank sections in the document

Code snippets containing console, class, or functions

legal or programming contexts

No Comments

No Known Activations