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. 16-GEMMASCOPE-RES-16K
    6. 6540
    Prev
    Next
    INDEX
    Explanations

    identifiers and variables that reference numerical or data-related concepts

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

    Code assignments and updates

    np_acts-logits-general · gemini-2.0-flash

    assignment to"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nakalista
    -1.04
    ]=="
    -0.86
    (!__
    -0.84
     {},
    
    -0.75
    LookAnd
    -0.75
    ']=='
    -0.74
     JAXBElement
    -0.74
     Савезне
    -0.74
    RenderAtEndOf
    -0.72
     \@
    -0.69
    POSITIVE LOGITS
     =
    0.83
     is
    0.62
     was
    0.53
     will
    0.52
     also
    0.47
    は
    0.47
     has
    0.44
     adalah
    0.43
    oa̍t
    0.43
    Addo
    0.41
    Activations Density 0.111%

    No Known Activations