© 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. 30655
    Prev
    Next
    INDEX
    Explanations

    quantitative measures or numerical values related to data processing or programming

    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
    275
    +0.14
    0.8%
    176
    +0.14
    0.8%
    395
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.14
    0.21
    395
    +0.14
    0.20
    36
    +0.14
    0.22
    Negative Logits
    ication
    -1.51
    itage
    -1.49
    ERTY
    -1.43
    orems
    -1.42
     APP
    -1.36
    nals
    -1.35
     Bak
    -1.34
    udson
    -1.32
    ruptcy
    -1.30
    ncia
    -1.29
    POSITIVE LOGITS
    ĥ½
    2.69
    ĭ
    2.67
    ľĵ
    2.63
    į
    2.60
    Ĥ
    2.53
    Ī
    2.50
    Ŀ
    2.45
    ↵↵↵
    2.45
           ↵
    2.45
    2.45
    Activations Density 8.474%

    No Known Activations