© 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. 26664
    Prev
    Next
    INDEX
    Explanations

    file paths or resource references

    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
    478
    +0.15
    0.8%
    17
    +0.14
    0.8%
    462
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    462
    +0.15
    0.04
    443
    +0.14
    0.04
    511
    +0.12
    0.03
    Negative Logits
    borne
    -1.82
    iest
    -1.69
    âĢIJ
    -1.55
    eed
    -1.52
    ochond
    -1.52
    ogenic
    -1.51
    genic
    -1.49
    à±į
    -1.45
    roscopic
    -1.42
    itively
    -1.41
    POSITIVE LOGITS
    ¿
    4.01
    ĥ½
    3.87
    ¦
    3.85
    ĭ
    3.84
    Ļª
    3.79
    Ĩ
    3.78
    ¾
    3.76
    ®
    3.72
    ĻĤ
    3.71
    Ŀ
    3.71
    Activations Density 0.174%

    No Known Activations