© 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. 29391
    Prev
    Next
    INDEX
    Explanations

    terms related to the process of compiling code

    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
    156
    +0.17
    1.0%
    162
    +0.13
    0.7%
    376
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    162
    +0.17
    0.02
    58
    +0.13
    0.01
    470
    +0.13
    0.01
    Negative Logits
    onas
    -1.68
    ajan
    -1.58
    ians
    -1.51
    ĸ
    -1.48
    lar
    -1.46
    ^](#
    -1.44
     peroxidase
    -1.37
    umab
    -1.37
    vir
    -1.36
    веÑĢ
    -1.35
    POSITIVE LOGITS
    pton
    1.76
    stance
    1.56
    anium
    1.46
    EMENT
    1.45
     error
    1.44
    herent
    1.39
     state
    1.38
    estry
    1.33
    emann
    1.33
    xture
    1.33
    Activations Density 0.083%

    No Known Activations