© 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 EvalsExportsAPI 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. 11400
    Prev
    Next
    INDEX
    Explanations

    phrases related to permissions and rights associated with software usage

    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
    21
    +0.13
    0.8%
    505
    +0.12
    0.7%
    508
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    411
    +0.13
    0.04
    69
    +0.12
    0.03
    505
    +0.11
    0.03
    Negative Logits
     gloves
    -1.48
     isolation
    -1.37
     opener
    -1.25
    less
    -1.25
     stret
    -1.24
     horns
    -1.21
     \[[
    -1.19
     favour
    -1.19
     strides
    -1.18
     stride
    -1.18
    POSITIVE LOGITS
     whom
    1.67
    duced
    1.64
    «
    1.64
    áĢº
    1.59
    áĢ¯
    1.54
    urs
    1.51
    áĢ
    1.48
     redistribute
    1.46
    glich
    1.45
    ublic
    1.43
    Activations Density 0.100%

    No Known Activations