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
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 3054
    Prev
    Next
    INDEX
    Explanations

    references to prior descriptions and discussions in the text

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

    phrases that reference previously mentioned content or cross-references to other sections.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    談社
    -0.52
     reten
    -0.49
    HideFlags
    -0.47
    aarrggbb
    -0.47
     GenerationType
    -0.43
     Autorizaciones
    -0.43
    Lähteet
    -0.42
    RTDA
    -0.41
    Ḷ
    -0.40
    tagext
    -0.40
    POSITIVE LOGITS
     elsewhere
    0.57
     herein
    0.53
     createState
    0.51
     in
    0.49
    WebVitals
    0.47
     below
    0.45
     CreateTagHelper
    0.44
     therein
    0.43
     earlier
    0.43
    SerializedSize
    0.42
    Activations Density 0.703%

    No Known Activations