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. 2-GEMMASCOPE-RES-16K
    6. 6868
    Prev
    Next
    INDEX
    Explanations

    patterns or structures in academic citations and references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 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
     виправивши
    -1.32
    AccessorTable
    -1.27
    MigrationBuilder
    -1.20
     AssemblyCulture
    -1.18
     disambiguazione
    -1.16
    AndEndTag
    -1.15
    MemoryWarning
    -1.15
     beginnetje
    -1.13
    ViewFeatures
    -1.13
    évaluateur
    -1.13
    POSITIVE LOGITS
     
    0.64
    ↵
    0.52
     at
    0.45
     in
    0.43
     i
    0.43
     a
    0.40
     (
    0.39
     -
    0.39
     on
    0.38
    ,
    0.38
    Activations Density 0.304%

    No Known Activations