© 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. 9356
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "Free" related to various contexts

    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
    437
    +0.20
    1.2%
    21
    +0.14
    0.8%
    316
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    437
    +0.20
    0.04
    218
    +0.14
    0.02
    503
    +0.11
    0.02
    Negative Logits
     themselves
    -1.57
     sorts
    -1.43
     harder
    -1.43
     save
    -1.41
     other
    -1.38
    sis
    -1.32
     consequence
    -1.31
     similarly
    -1.31
    apine
    -1.30
     controls
    -1.29
    POSITIVE LOGITS
    ª
    3.14
    Ļ
    3.09
    ³
    3.04
    ´
    2.99
    ½
    2.72
    ·¸
    2.70
    ı
    2.63
    İ
    2.61
    ĵ
    2.61
    ·
    2.59
    Activations Density 0.114%

    No Known Activations