© 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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 2948
    Prev
    Next
    INDEX
    Explanations

    actions or states related to confusion and disorganization

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

    Followed by a pronoun

    np_acts-logits-general · gemini-2.0-flash

    dealing with things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    expandindo
    -0.87
    StoryboardSegue
    -0.83
    Personendaten
    -0.81
     &___
    -0.79
    CppCodeGen
    -0.77
    +#+
    -0.76
    interopRequire
    -0.73
    Vidite
    -0.73
    onAttach
    -0.71
    OGND
    -0.68
    POSITIVE LOGITS
     things
    1.52
     everything
    1.19
    Things
    1.15
     Things
    1.11
     THINGS
    1.05
    everything
    1.00
    things
    0.98
     Everything
    0.95
    Everything
    0.92
     totul
    0.91
    Activations Density 0.237%

    No Known Activations