© 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. 18-GEMMASCOPE-RES-16K
    6. 2471
    Prev
    Next
    INDEX
    Explanations

    intentions and emotional responses related to personal beliefs and experiences

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

    Verbs followed by prepositions or conjunctions

    np_acts-logits-general · gemini-2.0-flash

    know, choose, earned, joined, love

    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
    BeginInit
    -0.60
    awtextra
    -0.50
    espèce
    -0.50
     "','
    -0.49
    BorderFactory
    -0.46
    species
    -0.45
    AntiForgeryToken
    -0.45
     kasarigan
    -0.45
    ">)</
    -0.44
    LabelTagHelper
    -0.44
    POSITIVE LOGITS
    portál
    0.69
     yourself
    0.67
     yourselves
    0.63
    contentPadding
    0.61
     ourselves
    0.61
     Himself
    0.61
     собі
    0.58
     himself
    0.58
     мәкал
    0.58
     herself
    0.58
    Activations Density 0.275%

    No Known Activations