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. 24-GEMMASCOPE-RES-131K
    6. 952
    Prev
    Next
    INDEX
    Explanations

    names of individuals, particularly in relation to their achievements and roles

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

    people's last names followed by verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    mor
    -0.36
     and
    -0.35
    ↵
    -0.33
     celor
    -0.33
     publice
    -0.33
     cang
    -0.32
     cassa
    -0.32
     reference
    -0.32
    Mor
    -0.32
    NameInMap
    -0.32
    POSITIVE LOGITS
    ########.
    0.65
    httphttps
    0.63
     صوتيه
    0.59
     bezeichneter
    0.58
     @"/
    0.58
    ſammen
    0.56
    ロウィン
    0.56
    
    0.55
    principalTable
    0.54
    󠀠
    0.54
    Activations Density 0.104%

    No Known Activations