© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 10512
    Prev
    Next
    INDEX
    Explanations

    phrases related to validity and correctness in various contexts

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

    words related to validity and legitimacy of claims or data.

    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
    antMatchers
    -0.40
    IInterface
    -0.39
    WithIOException
    -0.38
    枢
    -0.34
     HttpNotFound
    -0.31
    󠁢
    -0.31
    CppMethod
    -0.31
     handcu
    -0.30
    <_>
    -0.30
    NegativeButton
    -0.29
    POSITIVE LOGITS
     valid
    2.52
     validity
    2.20
    valid
    2.14
     Valid
    2.13
     VALID
    2.08
    Valid
    2.05
     invalid
    1.97
     Validity
    1.91
     isValid
    1.89
     valide
    1.76
    Activations Density 0.546%

    No Known Activations