© 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. 10997
    Prev
    Next
    INDEX
    Explanations

    terms related to software licensing and associated limitations

    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
    466
    +0.12
    0.7%
    397
    +0.11
    0.6%
    92
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.12
    0.02
    168
    +0.11
    0.02
    191
    +0.10
    0.02
    Negative Logits
    ļ
    -1.91
    ĵ
    -1.78
    ´
    -1.75
    stone
    -1.68
    IJ
    -1.65
    İ
    -1.64
    ı
    -1.63
    ĺ
    -1.59
    liner
    -1.59
    ĥ½
    -1.58
    POSITIVE LOGITS
    ftware
    2.00
    heit
    1.98
    anic
    1.61
    leep
    1.61
    rums
    1.57
    hips
    1.57
     grace
    1.56
     purposes
    1.52
     imposed
    1.48
    gue
    1.47
    Activations Density 0.171%

    No Known Activations