Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 10-RES-JB
    6. 721
    Prev
    Next
    INDEX
    Explanations

    instances of specific immune response types

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1842
    +0.11
    0.4%
    1036
    +0.10
    0.4%
    478
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.11
    0.03
    86
    +0.10
    0.03
    605
    +0.10
    0.01
    Negative Logits
    <bos>
    -1.49
    contentLoaded
    -0.97
    躇
    -0.89
    ScopeManager
    -0.73
    LayoutStyle
    -0.68
    Демографія
    -0.65
    titleMargin
    -0.64
    millimeters
    -0.61
     springfox
    -0.60
    HasAnnotation
    -0.60
    POSITIVE LOGITS
     unlaw
    1.25
     unwarran
    1.22
     impractica
    1.21
     ingrat
    1.13
     uninten
    1.09
     tolerably
    1.07
     disreg
    1.05
     impra
    1.05
     embodi
    1.03
     Longueur
    1.02
    Activations Density 0.349%

    No Known Activations