© 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. 31800
    Prev
    Next
    INDEX
    Explanations

    mentions of Microsoft and its activities or products

    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
    507
    +0.15
    0.9%
    212
    +0.15
    0.8%
    444
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.15
    0.23
    156
    +0.15
    0.23
    71
    +0.14
    0.23
    Negative Logits
    :`
    -1.84
    icable
    -1.57
    irected
    -1.50
    phasis
    -1.50
    ometh
    -1.50
     discipline
    -1.48
    most
    -1.43
     borne
    -1.40
     livelihood
    -1.39
    pled
    -1.38
    POSITIVE LOGITS
     cookies
    1.93
     Cookies
    1.89
     Suite
    1.74
     iOS
    1.59
    cdn
    1.57
     cookie
    1.54
     https
    1.52
     Cookie
    1.49
     twitter
    1.45
     logo
    1.45
    Activations Density 2.203%

    No Known Activations