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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13082
    Prev
    Next
    INDEX
    Explanations

    sections with significant gaps or breaks in content

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

    Code-related keywords: return, if, sc

    np_acts-logits-general · gemini-2.0-flash

    programming control flow and prior observations

    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
    <eos>
    -0.88
    ↵↵
    -0.84
    ↵↵↵
    -0.68
    ↵↵↵↵↵
    -0.56
    drawSprites
    -0.56
    ↵
    -0.55
    </h1>
    -0.53
    ↵↵↵↵
    -0.51
    LookAnd
    -0.50
    ↵↵↵↵↵↵↵
    -0.49
    POSITIVE LOGITS
    continue
    1.20
    goto
    0.85
    throw
    0.83
    break
    0.78
     continue
    0.74
     iſt
    0.69
     deswegen
    0.68
     deshalb
    0.68
    warnings
    0.67
     myſelf
    0.67
    Activations Density 0.094%

    No Known Activations