© 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. 25016
    Prev
    Next
    INDEX
    Explanations

    references to databases and database migration software

    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
    156
    +0.28
    1.6%
    379
    +0.13
    0.7%
    357
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.28
    0.03
    156
    +0.13
    0.02
    271
    +0.10
    0.01
    Negative Logits
    ager
    -1.58
     varying
    -1.55
    lying
    -1.51
    agem
    -1.47
     sexually
    -1.44
     studied
    -1.43
    apons
    -1.43
    aged
    -1.41
    itely
    -1.41
    off
    -1.40
    POSITIVE LOGITS
    plicity
    1.71
     engine
    1.70
     itself
    1.67
    DB
    1.58
    hton
    1.55
     Valley
    1.52
    )](
    1.52
    erals
    1.52
    lite
    1.47
    etica
    1.44
    Activations Density 0.103%

    No Known Activations