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. 22-GEMMASCOPE-RES-16K
    6. 8033
    Prev
    Next
    INDEX
    Explanations

    character strings or codes, likely related to programming or configuration settings

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

    Capital letters

    np_acts-logits-general · gemini-2.0-flash

    CV, FCVT

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Hoek
    -0.61
     Dickey
    -0.58
     Akhtar
    -0.58
    RTHOOK
    -0.57
     doInBackground
    -0.55
     Munster
    -0.55
     Richland
    -0.54
    FilterChain
    -0.54
    ngdoc
    -0.54
     Bartholo
    -0.53
    POSITIVE LOGITS
     Dav
    1.37
    Dav
    1.36
     Av
    1.35
     AV
    1.29
    Av
    1.28
     Cav
    1.27
     Cv
    1.26
     CV
    1.26
     Ev
    1.25
    dav
    1.24
    Activations Density 0.597%

    No Known Activations