© 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. 10168
    Prev
    Next
    INDEX
    Explanations

    phrases or terms associated with monetary or financial 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/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
    21
    +0.17
    1.0%
    434
    +0.14
    0.8%
    362
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.17
    0.01
    120
    +0.14
    0.03
    434
    +0.13
    0.02
    Negative Logits
     interest
    -1.88
     interests
    -1.68
    ez
    -1.56
    eper
    -1.49
    efore
    -1.43
     href
    -1.42
     minds
    -1.41
     respective
    -1.41
     unity
    -1.37
    OME
    -1.36
    POSITIVE LOGITS
    ½
    2.13
    ł
    2.11
    ¼
    2.08
    »
    2.06
    Ģ
    1.91
    Ł
    1.89
    ¿
    1.84
    ¡
    1.82
    ĵ
    1.81
    Īĺ
    1.77
    Activations Density 0.118%

    No Known Activations