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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 92928
    Prev
    Next
    INDEX
    Explanations

    references to spiritual beliefs and afterlife concepts

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

    preposition followed by noun or possessive

    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
    Matches
    -0.39
     Geste
    -0.38
    osexuality
    -0.38
    лару
    -0.37
    ⓧ
    -0.37
     kuva
    -0.36
     venu
    -0.35
    assen
    -0.34
     SAC
    -0.34
     AspNetCore
    -0.33
    POSITIVE LOGITS
    AddTagHelper
    0.60
    aarrggbb
    0.60
     sumpay
    0.58
    IVEREF
    0.56
     محفوظة
    0.56
    testify
    0.56
    GraphicsUnit
    0.52
     مرئيه
    0.52
    tilhe
    0.51
    BeginContext
    0.51
    Activations Density 0.001%

    No Known Activations