© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 2-GEMMASCOPE-RES-16K
    11. 4152
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Rot" and variations of rotation-related terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mybatisplus
    -0.53
     cohomology
    -0.42
     deoxy
    -0.42
    ="@+
    -0.40
    orten
    -0.40
    cabulary
    -0.39
     informée
    -0.39
    choso
    -0.38
    ╦
    -0.38
    ǧ
    -0.38
    POSITIVE LOGITS
     Rot
    0.87
     Rotating
    0.82
     rot
    0.82
    Rot
    0.81
     rotating
    0.81
     rotates
    0.81
    Rotating
    0.80
     rotate
    0.79
     rotated
    0.77
     rotational
    0.77
    Activations Density 0.485%

    No Known Activations