© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8120
    Prev
    Next
    INDEX
    Explanations

    references to expressions of opinion or statements of agreement

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

    Follows "I," "you," or "previously" and a verb of saying

    np_acts-logits-general · gemini-2.0-flash

    reporting speech or thoughts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <eos>
    -0.42
     war
    -0.42
    Record
    -0.42
    ума
    -0.41
    مد
    -0.40
    osp
    -0.40
     tud
    -0.39
    taf
    -0.39
     War
    -0.38
    </h2>
    -0.38
    POSITIVE LOGITS
     mentioned
    2.27
     said
    1.86
    mentioned
    1.84
     stated
    1.75
     Mentioned
    1.67
     mention
    1.63
     alluded
    1.61
    said
    1.59
     mentions
    1.53
     described
    1.47
    Activations Density 0.336%

    No Known Activations