© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 3999
    Prev
    Next
    INDEX
    Explanations

    terms related to earning, obtaining, and the concept of gaining something

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

    gaining or achieving something

    np_acts-logits-general · gemini-2.0-flash

    gaining status or rewards

    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
    existsSync
    -0.67
    HasForeignKey
    -0.56
    setInt
    -0.54
     nói
    -0.51
     acontecer
    -0.50
     rootReducer
    -0.49
    Happ
    -0.49
    cerita
    -0.48
     investigation
    -0.47
     Happ
    -0.47
    POSITIVE LOGITS
    Obtain
    1.03
     Obtaining
    1.02
     obtain
    0.98
     Obtain
    0.97
     obtains
    0.94
     obtained
    0.92
    obtained
    0.91
     Obtained
    0.90
     obtiene
    0.85
    acquire
    0.84
    Activations Density 0.151%

    No Known Activations