© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 2-RES-SM
    11. 30291
    Prev
    Next
    INDEX
    Explanations

    elements related to programming or code execution

    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
    455
    +0.17
    1.0%
    203
    +0.12
    0.7%
    491
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    432
    +0.17
    0.06
    53
    +0.12
    0.09
    77
    +0.12
    0.04
    Negative Logits
    oni
    -1.83
    ensing
    -1.80
    rant
    -1.79
    olin
    -1.77
    ert
    -1.71
    rief
    -1.68
    inger
    -1.64
    etica
    -1.62
    roelectric
    -1.61
    ieux
    -1.59
    POSITIVE LOGITS
    Ĥ
    2.28
    ĥ½
    2.08
    ij
    1.92
    ľĵ
    1.88
    Ĥ¬
    1.86
    Ģ
    1.86
    ¶
    1.84
      ↵
    1.84
    1.84
    1.84
    Activations Density 0.579%

    No Known Activations