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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24-GEMMASCOPE-RES-16K
    6. 4447
    Prev
    Next
    INDEX
    Explanations

    terms related to data security and privacy practices

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

    Encryption or security measures

    np_acts-logits-general · gemini-2.0-flash

    secure encryption and privacy

    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
    RenderAtEndOf
    -0.85
     mergeFrom
    -0.71
     CreateTagHelper
    -0.71
     ModelExpression
    -0.64
    AddTagHelper
    -0.60
    ⓧ
    -0.60
    pexpr
    -0.60
    országban
    -0.59
    ArgsConstructor
    -0.57
     nakalista
    -0.56
    POSITIVE LOGITS
     secure
    1.35
     encryption
    1.28
     Secure
    1.22
     encrypted
    1.16
    Secure
    1.15
     security
    1.15
    secure
    1.14
     secured
    1.13
     encrypt
    1.12
     securing
    1.09
    Activations Density 0.391%

    No Known Activations