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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 1-GEMMASCOPE-RES-16K
    6. 3130
    Prev
    Next
    INDEX
    Explanations

    phrases related to aggression or attacks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     незавершена
    -0.49
     fraude
    -0.44
     coû
    -0.43
    Datuak
    -0.43
     berlebihan
    -0.42
     estekak
    -0.40
    OGND
    -0.40
    Fail
    -0.40
     csal
    -0.40
    ember
    -0.39
    POSITIVE LOGITS
     attack
    1.45
     attacking
    1.44
     Attack
    1.38
    attack
    1.31
     attacked
    1.25
     ATTACK
    1.24
    Attacking
    1.20
     assault
    1.18
    Attack
    1.17
     attacks
    1.12
    Activations Density 0.346%

    No Known Activations