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-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 11485
    Prev
    Next
    INDEX
    Explanations

    phrases related to carrying out actions or procedures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     quarks
    -0.46
     millennium
    -0.46
    cordia
    -0.46
     licked
    -0.46
     Spinoza
    -0.45
    ruptcy
    -0.45
    Irrit
    -0.45
     Medvedev
    -0.45
     impres
    -0.45
     Drapeau
    -0.44
    POSITIVE LOGITS
    conducted
    1.21
     conducted
    1.16
     undertake
    1.12
     conducting
    1.11
     Conducted
    1.10
    performed
    1.09
     effectuer
    1.08
     undertook
    1.06
     conducts
    1.06
     undergone
    1.04
    Activations Density 0.428%

    No Known Activations