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. 24-GEMMASCOPE-RES-16K
    6. 8076
    Prev
    Next
    INDEX
    Explanations

    references to technology and its evolution

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

    "become" or its variants

    np_acts-logits-general · gemini-2.0-flash

    become

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    providedIn
    -0.57
    sidian
    -0.57
     تانيه
    -0.56
    Демографія
    -0.55
     gehad
    -0.54
     незавершена
    -0.52
     داشتند
    -0.48
    
    -0.48
     invokingState
    -0.48
     useHistory
    -0.46
    POSITIVE LOGITS
     become
    3.38
     becomes
    3.12
     becoming
    3.01
     became
    2.94
     Become
    2.92
    become
    2.91
    Become
    2.65
    becomes
    2.64
    becoming
    2.62
     Becomes
    2.60
    Activations Density 0.541%

    No Known Activations