© 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. 4-RES-SM
    11. 13048
    Prev
    Next
    INDEX
    Explanations

    initialization functions in class definitions

    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
    78
    +0.25
    1.5%
    397
    +0.16
    0.9%
    224
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.25
    0.00
    213
    +0.16
    0.03
    255
    +0.10
    -0.00
    Negative Logits
     suit
    -1.57
    asia
    -1.56
    autres
    -1.55
    idan
    -1.49
     desired
    -1.48
     closely
    -1.47
    gium
    -1.44
    Īĺ
    -1.43
     proportions
    -1.37
     prognosis
    -1.35
    POSITIVE LOGITS
    __(
    3.46
    __.
    3.33
    __)
    2.77
    __
    2.74
    __,
    2.70
    ___
    1.88
     __
    1.79
    method
    1.71
    rd
    1.70
    (){
    1.69
    Activations Density 0.206%

    No Known Activations