© 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. 4-RES-SM
    11. 12262
    Prev
    Next
    INDEX
    Explanations

    references to the term "man" and its variations in various contexts

    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
    8
    +0.13
    0.7%
    383
    +0.11
    0.6%
    197
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.13
    0.04
    8
    +0.11
    0.02
    494
    +0.11
    0.01
    Negative Logits
    ties
    -1.57
     ours
    -1.52
     theirs
    -1.47
    Ĥ
    -1.44
     true
    -1.41
    ¦
    -1.39
    ------------------------------------------------------------------------
    -1.39
     probability
    -1.39
     alumni
    -1.38
     members
    -1.37
    POSITIVE LOGITS
    hattan
    3.51
    uscript
    3.39
    ifest
    3.00
    ifold
    2.83
    iac
    2.74
    ufact
    2.74
    fred
    2.56
    ually
    2.53
    uel
    2.46
    agers
    2.46
    Activations Density 0.200%

    No Known Activations