© 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. 13018
    Prev
    Next
    INDEX
    Explanations

    references to the purpose of specific policies or legislative actions

    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
    379
    +0.11
    0.6%
    326
    +0.11
    0.6%
    434
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.11
    0.05
    407
    +0.11
    0.05
    108
    +0.10
    0.05
    Negative Logits
    Ļ
    -1.74
    ª
    -1.65
    ered
    -1.57
    ¾
    -1.48
    iator
    -1.45
     angle
    -1.44
    nd
    -1.41
    ĻĤ
    -1.37
    án
    -1.34
    «
    -1.34
    POSITIVE LOGITS
    operative
    1.40
    ductive
    1.33
     citing
    1.29
     exploration
    1.27
     Additionally
    1.24
     Collect
    1.24
    uding
    1.23
     hers
    1.23
    keeping
    1.21
     Comprehensive
    1.19
    Activations Density 0.507%

    No Known Activations