© 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. 1-RES-SM
    11. 24821
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate possession or relationships

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    98
    +0.19
    1.1%
    135
    +0.14
    0.8%
    30
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    30
    +0.19
    0.05
    98
    +0.14
    0.00
    457
    +0.12
    0.05
    Negative Logits
     prol
    -1.63
    dock
    -1.55
     docking
    -1.49
    mered
    -1.42
    hov
    -1.42
    ť
    -1.40
    cht
    -1.40
     mesh
    -1.37
     velocity
    -1.35
     behold
    -1.34
    POSITIVE LOGITS
    ¼
    4.48
    ł
    4.43
    ¿½
    4.43
    ĨĴ
    4.42
    Ń
    4.26
    ↵
    4.24
     ↵
    4.24
    ↵
    4.24
    ↵↵↵
    4.24
    ↵
    4.24
    Activations Density 0.343%

    No Known Activations