© 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. 30470
    Prev
    Next
    INDEX
    Explanations

    headings or sections indicating summaries or introductions in documents

    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
    71
    +0.13
    0.7%
    504
    +0.11
    0.6%
    271
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    131
    +0.13
    0.02
    298
    +0.11
    0.02
    491
    +0.11
    0.01
    Negative Logits
     remotely
    -1.72
     selling
    -1.51
     whatsoever
    -1.45
    except
    -1.43
     soit
    -1.43
     independent
    -1.43
    onym
    -1.40
     entirety
    -1.39
    cca
    -1.35
    dess
    -1.35
    POSITIVE LOGITS
    ł
    1.88
    ij
    1.85
    ¢
    1.83
     BGCOLOR
    1.81
    ¼
    1.77
    £
    1.71
    ·¸
    1.65
    ŀ
    1.64
    ¡
    1.63
    ³
    1.55
    Activations Density 0.044%

    No Known Activations