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. 103118
    Prev
    Next
    INDEX
    Explanations

    phrases related to purchases and assistance in various contexts

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

    Verbs used as main actions in sentences, particularly those related to transactions (purchase), assistance (assist), or physical actions.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
     المعيارى
    -0.76
    +#+#
    -0.68
    Belly
    -0.64
     Houſe
    -0.63
     Anſ
    -0.63
    󠁮
    -0.63
    RectangleBorder
    -0.61
    Everything
    -0.61
     bureauc
    -0.59
     Theſe
    -0.59
    POSITIVE LOGITS
     iż
    0.66
     lecz
    0.63
     upon
    0.62
     utilizing
    0.61
     utilising
    0.60
     pertanto
    0.58
    maktadır
    0.57
    mektedir
    0.57
     whilst
    0.56
     effettuare
    0.54
    Activations Density 0.294%

    No Known Activations