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. 11226
    Prev
    Next
    INDEX
    Explanations

    phrases related to error handling and command execution in programming contexts

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

    Error messages or invalid states

    np_acts-logits-general · gemini-2.0-flash

    invalid command or URL

    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
    SourceChecksum
    -0.70
    parsedMessage
    -0.70
    __*/
    -0.64
    +#+#
    -0.59
     nawr
    -0.58
     defaultstate
    -0.57
    AccessorTable
    -0.55
    θεια
    -0.55
     sparing
    -0.54
     endforeach
    -0.51
    POSITIVE LOGITS
     invalid
    1.05
     Invalid
    0.94
     incorrect
    0.94
    invalid
    0.93
     cannot
    0.92
    Invalid
    0.89
     failed
    0.89
     must
    0.88
     Cannot
    0.88
     wrong
    0.87
    Activations Density 0.378%

    No Known Activations