© 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. 4-RES-SM
    11. 9032
    Prev
    Next
    INDEX
    Explanations

    significant numerical values or identifiers related to scientific or technical content

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    271
    +0.13
    0.8%
    306
    +0.11
    0.7%
    445
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.13
    0.03
    255
    +0.11
    0.00
    477
    +0.11
    0.04
    Negative Logits
    entary
    -1.56
    ctions
    -1.52
    uates
    -1.39
    inates
    -1.38
    gery
    -1.37
    ported
    -1.35
     accord
    -1.30
     site
    -1.29
    lectual
    -1.28
     attendance
    -1.27
    POSITIVE LOGITS
    ĥ½
    2.89
    ³
    2.51
    Ļ
    2.43
    ļ
    2.35
    ²
    2.27
    Į
    2.27
    ¸
    2.27
    Ķ
    2.19
    ´
    2.16
    ĺ
    2.16
    Activations Density 0.204%

    No Known Activations