© 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-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 14269
    Prev
    Next
    INDEX
    Explanations

    expressions of doubt or negative opinions about a subject

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

    the contraction "wouldn't" and words that follow it expressing negative opinions or refusals.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    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
    bootstrapcdn
    -0.49
    MethodManager
    -0.49
     zoude
    -0.47
    StoryboardSegue
    -0.46
    basicConfig
    -0.46
    LEncoder
    -0.45
    المكان
    -0.45
    AnimationsModule
    -0.44
     Мексичка
    -0.44
     claramente
    -0.44
    POSITIVE LOGITS
     tolerate
    0.71
     tolerated
    0.69
     EVER
    0.66
     accept
    0.65
     toler
    0.63
    toler
    0.61
     ever
    0.59
     survive
    0.58
     ACCEPT
    0.55
     budge
    0.53
    Activations Density 0.216%

    No Known Activations