© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 15067
    Prev
    Next
    INDEX
    Explanations

    terms related to limitations and restrictions

    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
    78
    +0.19
    1.1%
    240
    +0.13
    0.8%
    419
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.19
    0.01
    78
    +0.13
    -0.00
    240
    +0.11
    0.00
    Negative Logits
    chle
    -1.40
     {};
    -1.34
    otrophic
    -1.33
     commune
    -1.29
    ctic
    -1.27
     preferably
    -1.26
    average
    -1.21
    flix
    -1.21
    veau
    -1.20
    yne
    -1.19
    POSITIVE LOGITS
    Į
    4.22
    Ĵ
    3.87
    µ
    3.87
    ĥ
    3.86
    Ī
    3.79
    ©
    3.79
    ·¸
    3.66
    ŀ
    3.62
    ¿½
    3.61
    §
    3.59
    Activations Density 0.070%

    No Known Activations