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. 11-GEMMASCOPE-RES-16K
    6. 13635
    Prev
    Next
    INDEX
    Explanations

    quantitative data and numerical relationships regarding various scientific and mathematical concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     autorytatywna
    -1.09
     pinulongan
    -0.91
    Geplaatst
    -0.80
    InjectAttribute
    -0.80
     >=",
    -0.80
     beginnetje
    -0.78
     esternos
    -0.77
    CGRectMake
    -0.77
    EndTag
    -0.76
    extAlignment
    -0.75
    POSITIVE LOGITS
     overall
    0.59
     totalt
    0.48
     Overall
    0.48
    Overall
    0.47
     totale
    0.44
    overall
    0.44
     sanitaires
    0.43
     per
    0.43
     total
    0.41
     than
    0.40
    Activations Density 0.637%

    No Known Activations