© 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 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. 10546
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to testing, evaluation, and measurement standards 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/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
    +0.18
    1.1%
    47
    +0.15
    0.9%
    354
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    340
    +0.18
    0.07
    441
    +0.15
    -0.01
    408
    +0.13
    0.08
    Negative Logits
    ĥ½
    -2.46
    ľ
    -2.37
    Į
    -2.09
    ²
    -1.99
    «
    -1.98
    ĺ
    -1.97
    ¡
    -1.96
    ľĵ
    -1.92
    Ŀ
    -1.92
    ł
    -1.81
    POSITIVE LOGITS
    ively
    1.64
    enix
    1.57
    ographically
    1.51
    oj
    1.43
    nt
    1.38
    ently
    1.37
    only
    1.36
    rog
    1.32
    lessly
    1.31
    ocardi
    1.31
    Activations Density 3.953%

    No Known Activations