INDEX

Explanations

system-related error signals and codes

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

legal or technical errors

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

 randomised

-0.46

sidemargin

-0.45

calciatore

-0.43

 Basal

-0.43

ppincott

-0.43

etton

-0.42

ghijklmnop

-0.42

anwal

-0.42

OOTDTY

-0.41

 Sunset

-0.40

POSITIVE LOGITS

SequentialGroup

0.53

 Италијани

0.52

AsUp

0.46

GenerationType

0.42

@"/

0.42

ViewFeatures

0.42

ModelAdmin

0.40

dafx

0.40

findpost

0.39

 AssemblyCompany

0.39

Activations Density 0.129%

system-related error signals and codes

legal or technical errors

No Comments

No Known Activations