© 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. 2-RES-SM
    11. 30673
    Prev
    Next
    INDEX
    Explanations

    comparative phrases and expressions indicating differences or relationships between subjects

    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
    265
    +0.13
    0.7%
    18
    +0.12
    0.7%
    240
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.13
    0.03
    284
    +0.12
    0.03
    240
    +0.12
    0.00
    Negative Logits
     reputation
    -1.89
     likes
    -1.69
     lots
    -1.54
     popularity
    -1.46
     short
    -1.41
     success
    -1.40
     windows
    -1.39
     wal
    -1.36
    uities
    -1.35
     fuck
    -1.33
    POSITIVE LOGITS
    DG
    1.51
    AFFIRMED
    1.49
    ched
    1.45
     conferred
    1.40
    /~
    1.37
     employed
    1.36
    gaard
    1.35
    <|outofrange|>
    1.35
    ↵
    1.35
    ↵
    1.35
    Activations Density 0.370%

    No Known Activations