© 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. 27052
    Prev
    Next
    INDEX
    Explanations

    patterns in structured data or markup language

    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
    419
    +0.15
    0.8%
    192
    +0.13
    0.7%
    326
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    330
    +0.15
    0.10
    131
    +0.13
    0.11
    475
    +0.12
    0.05
    Negative Logits
    Ļª
    -2.48
    £
    -2.12
     respectively
    -2.07
    ĭ
    -2.06
    IJ
    -1.94
    ±
    -1.94
    ĻĤ
    -1.84
    ı
    -1.83
    »¿
    -1.77
    ĵ
    -1.77
    POSITIVE LOGITS
    vere
    1.84
     tension
    1.50
     expans
    1.42
    Generated
    1.34
    helm
    1.27
    odynamic
    1.26
    degree
    1.26
    piece
    1.21
    sided
    1.21
     shoulders
    1.21
    Activations Density 4.391%

    No Known Activations