Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3542
    Prev
    Next
    INDEX
    Explanations

    expressions of strong frustration or anger

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Expletives (fuck, shit, damn)

    np_acts-logits-general · gemini-2.0-flash

    Fuck and shit

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    parsedMessage
    -0.49
     ligiloj
    -0.49
     kasarigan
    -0.45
    requent
    -0.44
     సౌకర్య
    -0.44
    曖昧さ回避
    -0.43
     mergeFrom
    -0.42
    Personendaten
    -0.41
    ItemBackground
    -0.40
     bereit
    -0.40
    POSITIVE LOGITS
     shit
    0.83
    fuck
    0.81
     fucking
    0.81
     fuck
    0.78
    Fuck
    0.77
     Shit
    0.77
    Shit
    0.77
    shit
    0.76
    fucking
    0.76
     SHIT
    0.75
    Activations Density 0.222%

    No Known Activations