INDEX

Explanations

structured data formats and training data references

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

data structures with numbers

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

MLLoader

-0.92

 ब्रेकडाउन

-0.73

setVerticalGroup

-0.69

 BoxDecoration

-0.66

 pinulongan

-0.64

TemporalType

-0.63

AnimationsModule

-0.62

lapsingToolbar

-0.60

⤹

-0.60

 оригіналу

-0.59

POSITIVE LOGITS

 memory

0.40

 Learning

0.38

Tabela

0.37

 Table

0.35

du

0.35

TableId

0.34

 museum

0.34

InlineData

0.34

 Jake

0.34

InfoList

0.33

Activations Density 0.230%

structured data formats and training data references

data structures with numbers

No Comments

No Known Activations