© 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 EvalsExportsAPI 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. 26404
    Prev
    Next
    INDEX
    Explanations

    structured coding syntax or markup indicators

    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
    478
    +0.17
    0.9%
    196
    +0.15
    0.8%
    77
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.17
    0.03
    31
    +0.15
    0.03
    84
    +0.14
    0.02
    Negative Logits
    ftime
    -1.80
    ikov
    -1.74
     features
    -1.68
    oton
    -1.64
    ery
    -1.53
    coat
    -1.51
    itzer
    -1.49
    key
    -1.49
    itter
    -1.48
    mass
    -1.43
    POSITIVE LOGITS
    theless
    1.72
     mind
    1.65
    fraid
    1.50
    ~\
    1.49
    selves
    1.48
     forth
    1.47
    ~^
    1.47
     imagination
    1.43
     doub
    1.42
    ertainly
    1.41
    Activations Density 0.192%

    No Known Activations