© 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. 30639
    Prev
    Next
    INDEX
    Explanations

    references to recent advancements and research in various fields

    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
    326
    +0.13
    0.7%
    418
    +0.12
    0.7%
    459
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    418
    +0.13
    -0.00
    33
    +0.12
    0.08
    365
    +0.12
    0.06
    Negative Logits
    Īĺ
    -3.20
    Ľ
    -2.91
    ¹
    -2.70
    ĻĤ
    -2.68
    ĨĴ
    -2.67
    §
    -2.63
    ³
    -2.58
    ↵
    -2.41
    ↵    ↵
    -2.41
    ↵
    -2.41
    POSITIVE LOGITS
    Than
    1.69
    Theorem
    1.55
     analogues
    1.52
    olev
    1.45
    methods
    1.45
     technologies
    1.44
    method
    1.44
    Proposition
    1.44
    avelength
    1.44
     resear
    1.42
    Activations Density 1.136%

    No Known Activations