INDEX

Explanations

numerical representations and mathematical expressions

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

lists of numbers and symbols

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

 صوتيه

-0.42

PARATOR

-0.42

oa̍t

-0.40

 thorough

-0.40

➋

-0.37

PERATURE

-0.37

dem

-0.36

Dem

-0.36

原始内容存档于

-0.35

 hardly

-0.34

POSITIVE LOGITS

ItemBackground

0.61

outWeight

0.61

aarrggbb

0.57

parsedMessage

0.53

GEBURTS

0.53

ViewFeatures

0.52

ChildScrollView

0.49

LabelTagHelper

0.49

enderror

0.46

DeleteBehavior

0.46

Activations Density 2.300%

numerical representations and mathematical expressions

lists of numbers and symbols

No Comments

No Known Activations