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. 30462
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and button-related elements in the text

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

    code snippets and commands

    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
    \{\\
    -0.52
     modelAndView
    -0.49
    bootstrapcdn
    -0.48
    tvguidetime
    -0.48
    balleur
    -0.48
     Taktlose
    -0.46
    ectoria
    -0.45
    ({})
    -0.44
    Clik
    -0.44
    SpringRunner
    -0.44
    POSITIVE LOGITS
    TabIndex
    0.93
     =
    0.57
     StatelessWidget
    0.46
    setVerticalGroup
    0.42
     useRef
    0.42
    GEBURTSDATUM
    0.42
     mensuales
    0.41
    >=</
    0.41
     ^=
    0.40
     hint
    0.39
    Activations Density 0.000%

    No Known Activations