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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 24194
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly exclamation and question marks, as well as special characters used in expressions

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

    !important or !global

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -------
    -0.54
     때문에
    -0.49
     насељу
    -0.48
    เย็น
    -0.48
     Assessments
    -0.47
     Contacto
    -0.47
    发表于
    -0.46
    TextBoxColumn
    -0.45
     Ambiental
    -0.44
     Geduld
    -0.44
    POSITIVE LOGITS
    ghijklmnop
    0.76
    arangay
    0.71
    ésult
    0.65
    copg
    0.59
     مرئيه
    0.56
    ulongan
    0.53
    viewDid
    0.53
    UALA
    0.52
    Vanjske
    0.52
    stdc
    0.51
    Activations Density 0.023%

    No Known Activations