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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 7836
    Prev
    Next
    INDEX
    Explanations

    instances of the word "don't" or its variants

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

    "I don't" or "don't"

    np_acts-logits-general · gemini-2.0-flash

    i don't think/know/see

    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
    GOTREF
    -0.75
    PyExc
    -0.61
    __':
    -0.61
    HostException
    -0.59
    Geplaatst
    -0.57
    rrggbb
    -0.57
    InitStruct
    -0.55
     Utama
    -0.55
    BASEPATH
    -0.54
     "..\..\..\
    -0.53
    POSITIVE LOGITS
     recall
    0.78
    recall
    0.65
     envy
    0.62
     Recall
    0.61
     think
    0.57
     how
    0.55
     remember
    0.54
     know
    0.54
     vorrei
    0.54
     why
    0.52
    Activations Density 0.124%

    No Known Activations