© 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. 4-RES-SM
    11. 9775
    Prev
    Next
    INDEX
    Explanations

    words related to legal terms and processes

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    156
    +0.43
    2.6%
    277
    +0.13
    0.8%
    307
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    277
    +0.43
    0.04
    198
    +0.13
    0.10
    253
    +0.10
    0.11
    Negative Logits
    eks
    -1.56
    quo
    -1.56
    nai
    -1.55
    ked
    -1.52
    heet
    -1.50
    aws
    -1.36
    Ñħ
    -1.34
    STRA
    -1.34
    Äij
    -1.33
    ÑĥÑĢ
    -1.32
    POSITIVE LOGITS
    »¿
    3.15
    Ļª
    3.14
    ·¸
    3.11
    £
    2.92
    ĻĤ
    2.91
    ¿
    2.90
    Ń
    2.83
    ´
    2.78
    µ
    2.68
    ı
    2.67
    Activations Density 1.523%

    No Known Activations