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-2-2B
    4. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 16246
    Prev
    Next
    INDEX
    Explanations

    various forms of underscores and hyphens, possibly indicative of formatting elements or named entities in structured data

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

    Filenames or IDs with dashes and numbers

    np_acts-logits-general · gemini-2.0-flash

    image or table references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RUnlock
    -0.71
    AndEndTag
    -0.66
     useDispatch
    -0.63
    Kjelder
    -0.62
    PhysRev
    -0.61
    imageshack
    -0.59
     Commencez
    -0.56
    addPreferredGap
    -0.56
    basicConfig
    -0.55
    addGap
    -0.54
    POSITIVE LOGITS
    onViewCreated
    0.51
    ніципа
    0.49
    CPtr
    0.48
    ως
    0.47
     
    0.46
    SharedCtor
    0.46
    AMAN
    0.46
     Humberto
    0.44
     Tavares
    0.44
     English
    0.44
    Activations Density 0.369%

    No Known Activations