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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 131070
    Prev
    Next
    INDEX
    Explanations

    references to personal relationships and connections, particularly emphasizing friendship and familial ties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    ⓘ
    -0.40
    parseDouble
    -0.40
    ariConfig
    -0.39
    ⤹
    -0.36
    Skocz
    -0.35
    extAlignment
    -0.34
    XMLSchema
    -0.34
    <?>>
    -0.33
     la
    -0.33
    PyErr
    -0.31
    POSITIVE LOGITS
     friends
    0.68
    mmate
    0.61
     colleagues
    0.61
    Personendaten
    0.60
     colleague
    0.60
     collega
    0.59
    friends
    0.58
     friend
    0.58
     acquaintances
    0.58
     sorella
    0.58
    Activations Density 0.061%

    No Known Activations