© 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. 13010
    Prev
    Next
    INDEX
    Explanations

    significant structural elements of a programming code or document

    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
    410
    +0.18
    1.0%
    59
    +0.13
    0.7%
    111
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    163
    +0.18
    0.02
    410
    +0.13
    0.03
    165
    +0.10
    0.02
    Negative Logits
    hab
    -1.62
    ypse
    -1.56
    former
    -1.55
    Inflater
    -1.55
     reminiscent
    -1.54
     television
    -1.53
    artment
    -1.50
     transporter
    -1.48
     contral
    -1.46
    olean
    -1.46
    POSITIVE LOGITS
    ģ
    2.03
    ±
    2.02
    ³
    1.99
    Īĺ
    1.88
    ²
    1.88
    ·¸
    1.88
    ¾
    1.87
    ¥
    1.85
    ĻĤ
    1.83
    ¬
    1.82
    Activations Density 0.057%

    No Known Activations