© 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. 1-RES-SM
    11. 28598
    Prev
    Next
    INDEX
    Explanations

    phrases related to XML and document structure

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    69
    +0.10
    0.6%
    252
    +0.10
    0.6%
    253
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    253
    +0.10
    0.02
    81
    +0.10
    0.02
    252
    +0.09
    0.01
    Negative Logits
     suspicion
    -1.65
    inations
    -1.56
    ...](
    -1.49
     mercy
    -1.48
     destiny
    -1.45
     sudden
    -1.45
     blame
    -1.44
     accidents
    -1.44
    ACE
    -1.43
     Parenthood
    -1.42
    POSITIVE LOGITS
    enthal
    2.10
    ercase
    1.87
    ÅĽci
    1.82
    nyder
    1.74
    chet
    1.69
    ých
    1.65
    ÃŃf
    1.63
    ian
    1.59
    ENTIAL
    1.57
    wald
    1.57
    Activations Density 0.069%

    No Known Activations