Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 4578
    Prev
    Next
    INDEX
    Explanations

    terms related to research methodologies and their applications in social sciences

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

    preceding theories or theoretical frameworks

    np_acts-logits-general · gemini-2.0-flash

    theories and concepts

    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
     ***!
    -0.59
     onPostExecute
    -0.57
    AllAfrica
    -0.57
    istoitu
    -0.53
     nakalista
    -0.53
     VIAF
    -0.51
     muualla
    -0.50
     дописавши
    -0.50
    τως
    -0.50
     Allein
    -0.49
    POSITIVE LOGITS
     theory
    1.28
     theories
    1.26
    theory
    1.07
     theorists
    1.01
     Theory
    0.98
     Theories
    0.94
     teoría
    0.93
     concepts
    0.92
     теории
    0.92
    Theory
    0.91
    Activations Density 0.336%

    No Known Activations