© 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. 13200
    Prev
    Next
    INDEX
    Explanations

    legal terms and references to court cases

    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
    78
    +0.13
    0.8%
    388
    +0.11
    0.7%
    358
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.13
    -0.02
    78
    +0.11
    -0.02
    388
    +0.11
    0.01
    Negative Logits
    tons
    -1.50
     consortium
    -1.40
    giving
    -1.34
    typeof
    -1.34
    iba
    -1.32
     Mum
    -1.32
    ools
    -1.29
     jazz
    -1.27
     bass
    -1.26
     architecture
    -1.24
    POSITIVE LOGITS
    IJ
    2.93
    »
    2.77
    Ĩ
    2.71
    ĵ
    2.70
    ľĵ
    2.69
    ĥ½
    2.68
    Ī
    2.67
    ·
    2.67
    ¢
    2.63
    ↵
    2.62
    Activations Density 0.516%

    No Known Activations