© 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. 1-RES-SM
    11. 29312
    Prev
    Next
    INDEX
    Explanations

    specific terms related to formal documents, legal context, and academic 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    438
    +0.15
    0.8%
    224
    +0.12
    0.6%
    424
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    297
    +0.15
    0.01
    453
    +0.12
    0.01
    262
    +0.11
    0.01
    Negative Logits
     awake
    -1.68
    ť
    -1.49
    ined
    -1.46
     intox
    -1.46
     Into
    -1.43
     sed
    -1.39
     hooked
    -1.39
     Ethiop
    -1.36
    wers
    -1.32
     ye
    -1.32
    POSITIVE LOGITS
    £
    3.31
    ¢
    2.93
    ł
    2.91
    »¿
    2.76
    ¨
    2.76
    ↵
    2.61
    č↵
    2.61
    Â
    2.61
    2.61
    2.61
    Activations Density 0.095%

    No Known Activations