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. 4358
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter "V" in various contexts

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

    "V" followed by a period or other letters

    np_acts-logits-general · gemini-2.0-flash

    V. abbreviations or followed by common names

    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
    SourceChecksum
    -0.85
    
    -0.81
    ioutil
    -0.69
    AndEndTag
    -0.65
    enumi
    -0.64
     ProtoMessage
    -0.61
     defaultstate
    -0.61
     gyhoeddwyd
    -0.56
    hyrchwyd
    -0.55
    __(/*!
    -0.54
    POSITIVE LOGITS
     iload
    0.51
    PositiveButton
    0.49
     basi
    0.49
    AddTagHelper
    0.47
    locking
    0.46
    alium
    0.45
    توضیحات
    0.44
    ề
    0.44
    batore
    0.44
     Coalition
    0.43
    Activations Density 0.170%

    No Known Activations