© 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. 13052
    Prev
    Next
    INDEX
    Explanations

    references to semi-structured data or documents

    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
    59
    +0.12
    0.7%
    350
    +0.11
    0.6%
    442
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.12
    0.06
    338
    +0.11
    0.03
    59
    +0.10
    0.03
    Negative Logits
    ľĵ
    -1.87
    Ĥ
    -1.83
    breviations
    -1.75
    ¿½
    -1.52
    »
    -1.52
    Į
    -1.51
    claimed
    -1.49
    ĻĤ
    -1.46
    ½
    -1.45
    Ĺ
    -1.44
    POSITIVE LOGITS
    ersion
    1.54
     stroll
    1.51
     angle
    1.50
    ures
    1.46
     jog
    1.46
     quarters
    1.41
     moonlight
    1.36
     Crus
    1.32
     hats
    1.29
     turns
    1.28
    Activations Density 0.892%

    No Known Activations