© 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. 9625
    Prev
    Next
    INDEX
    Explanations

    legal case identifiers and related terms

    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
    263
    +0.27
    1.6%
    186
    +0.19
    1.2%
    369
    +0.19
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.27
    0.28
    156
    +0.19
    0.28
    71
    +0.19
    0.28
    Negative Logits
     pores
    -1.59
    scriptions
    -1.53
     depths
    -1.49
     conform
    -1.48
     parte
    -1.45
    ibraries
    -1.41
    ifacts
    -1.39
    esis
    -1.35
     debris
    -1.35
     Papers
    -1.35
    POSITIVE LOGITS
    ĻĤ
    2.94
    »¿
    2.70
    ĨĴ
    2.69
    Ļª
    2.56
    §
    2.55
    ī
    2.53
    ©
    2.52
    ¶
    2.51
    ±
    2.44
    ı
    2.43
    Activations Density 4.342%

    No Known Activations