© 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. 9902
    Prev
    Next
    INDEX
    Explanations

    legal citations and references to court cases

    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
    255
    +0.13
    0.7%
    175
    +0.12
    0.7%
    55
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.13
    0.00
    84
    +0.12
    0.02
    326
    +0.11
    0.02
    Negative Logits
    ctors
    -1.58
    etheless
    -1.52
    verter
    -1.49
    iginally
    -1.42
     doubt
    -1.38
    lessly
    -1.35
     another
    -1.34
     normally
    -1.34
    poser
    -1.32
    verts
    -1.31
    POSITIVE LOGITS
    ´
    5.35
    ¥
    5.15
    Ŀ
    5.08
    °
    5.02
    Ĩ
    5.02
    £
    5.01
    ¾
    4.99
    ³
    4.97
    ¦
    4.94
    ¯
    4.92
    Activations Density 0.064%

    No Known Activations