INDEX

Explanations

terms related to networks and modular structures

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

dividing into sub-parts or units

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

SequentialGroup

-0.50

XmlAccessType

-0.41

ngdoc

-0.38

odyne

-0.37

dur

-0.36

disposing

-0.33

readFileSync

-0.33

fromCharCode

-0.32

ط

-0.32

PreExecute

-0.32

POSITIVE LOGITS

GraphicsUnit

0.54

 isolato

0.54

 transfieras

0.54

一个个

0.54

 neighbourhoods

0.53

 terpisah

0.52

 isolé

0.52

 jednot

0.51

IsMutable

0.51

⟬

0.51

Activations Density 0.104%

terms related to networks and modular structures

dividing into sub-parts or units

No Comments

No Known Activations

terms related to networks and modular structures

dividing into sub-parts or units

No Comments

No Known Activations