Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 5-GEMMASCOPE-RES-16K
    6. 11583
    Prev
    Next
    INDEX
    Explanations

    references to "raw" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ngdoc
    -1.00
     <=",
    -0.92
    tvguidetime
    -0.92
     AssemblyVersion
    -0.81
    ///</
    -0.80
     épis
    -0.79
     comigo
    -0.78
     "..\..\
    -0.78
    matchCondition
    -0.77
    ніципа
    -0.76
    POSITIVE LOGITS
     Raw
    2.49
     raw
    2.40
    Raw
    2.32
     RAW
    2.13
    raw
    2.10
    RAW
    1.99
    getRaw
    1.52
    RawData
    1.41
    rawData
    1.41
     cruda
    1.26
    Activations Density 0.059%

    No Known Activations