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

    Neuronpedia

    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. 10847
    Prev
    Next
    INDEX
    Explanations

    references to cells in a data structure or programming context

    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
    156
    +0.23
    1.4%
    483
    +0.12
    0.7%
    269
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    446
    +0.23
    0.01
    164
    +0.12
    0.02
    26
    +0.10
    0.01
    Negative Logits
    ness
    -1.70
    cause
    -1.55
     cause
    -1.54
     behalf
    -1.54
     suspicion
    -1.45
     maternal
    -1.45
    use
    -1.45
     severe
    -1.43
    ĥ½
    -1.42
    ½
    -1.35
    POSITIVE LOGITS
    phone
    2.30
     phone
    2.19
     phones
    2.11
    phones
    1.91
    ette
    1.88
    ettes
    1.84
    inal
    1.61
     Phone
    1.60
    ulose
    1.57
    arez
    1.57
    Activations Density 0.144%

    No Known Activations