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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 28-GEMMASCOPE-RES-16K
    6. 4058
    Prev
    Next
    INDEX
    Explanations

    assertions related to certainty and evidence in legal contexts

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

    certainty or lack of doubt

    np_acts-logits-general · gemini-2.0-flash

    beyond reasonable doubt

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CppCodeGen
    -0.65
     AssemblyProduct
    -0.49
    endpush
    -0.44
    inSlope
    -0.39
    ankind
    -0.36
    swag
    -0.35
     tartalomajánló
    -0.35
    HomeAsUpEnabled
    -0.35
    toThrow
    -0.34
    peutic
    -0.34
    POSITIVE LOGITS
     doubt
    0.97
     undeniable
    0.95
     indisputable
    0.92
     undoubted
    0.90
     undoubtedly
    0.77
     undisputed
    0.76
     unquestionably
    0.76
     Undoubtedly
    0.75
    doubt
    0.73
     unquestion
    0.72
    Activations Density 0.306%

    No Known Activations