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-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15680
    Prev
    Next
    INDEX
    Explanations

    references to future developments or assessments

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

    Following "in", "than", or "for" referring to time

    np_acts-logits-general · gemini-2.0-flash

    future, past, previous, later

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     الحره
    -0.60
    何でも
    -0.47
    なんでも
    -0.47
    Otherwise
    -0.46
     Otherwise
    -0.45
    addPreferredGap
    -0.45
    LabelTagHelper
    -0.45
    nóstico
    -0.44
    Secondly
    -0.43
     ſta
    -0.43
    POSITIVE LOGITS
     previous
    1.89
    previous
    1.63
     past
    1.52
    Previous
    1.52
     future
    1.52
     Previous
    1.45
     subsequent
    1.38
    past
    1.34
     recent
    1.32
    future
    1.30
    Activations Density 0.495%

    No Known Activations