© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 9850
    Prev
    Next
    INDEX
    Explanations

    programming language syntax, particularly for defining and managing data types and variables

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    153
    +0.22
    1.3%
    320
    +0.18
    1.0%
    419
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.22
    0.14
    157
    +0.18
    0.13
    320
    +0.17
    0.10
    Negative Logits
    /*
    -1.38
    orship
    -1.36
    azine
    -1.36
     Mrs
    -1.30
    ctomy
    -1.30
     Mr
    -1.30
    #:
    -1.28
     âĢ¢
    -1.28
     rounds
    -1.28
    erto
    -1.27
    POSITIVE LOGITS
    argin
    1.71
    Īĺ
    1.63
    anges
    1.61
    keit
    1.55
    romy
    1.39
    icum
    1.38
    ems
    1.35
    cyl
    1.35
    apes
    1.34
    asio
    1.31
    Activations Density 2.256%

    No Known Activations