© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 14021
    Prev
    Next
    INDEX
    Explanations

    references to proteins and their associated functions or characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    283
    +0.14
    0.8%
    289
    +0.13
    0.7%
    442
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.14
    0.02
    289
    +0.13
    0.01
    1
    +0.12
    0.01
    Negative Logits
     happen
    -1.57
     Medicare
    -1.47
     suppose
    -1.46
     intention
    -1.45
    ordable
    -1.44
    erre
    -1.39
    omething
    -1.39
    rians
    -1.39
    empre
    -1.38
    imony
    -1.37
    POSITIVE LOGITS
    uria
    2.81
    aceous
    2.50
     kinase
    2.43
     synthesis
    2.43
     kinases
    2.33
     complexes
    2.27
     tyrosine
    2.18
     binding
    2.12
     phosphatase
    2.07
     folding
    2.01
    Activations Density 0.051%

    No Known Activations