INDEX

Explanations

alphanumeric codes or identifiers that may indicate links or references

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

urls and links

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

ssymb

-0.35

 ange

-0.35

 سرا

-0.35

 nonetheless

-0.35

 increasingly

-0.35

+#+#

-0.35

mord

-0.35

 closely

-0.35

 honor

-0.34

 servire

-0.34

POSITIVE LOGITS

 AssemblyCompany

0.62

ViewImports

0.55

parsedMessage

0.54

 surla

0.54

حوالہ

0.53

sidemargin

0.50

 nonUne

0.49

Personensuche

0.49

Rohy

0.48

aarrggbb

0.48

Activations Density 0.016%

alphanumeric codes or identifiers that may indicate links or references

urls and links

No Comments

No Known Activations