© 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. 2-RES-SM
    11. 10492
    Prev
    Next
    INDEX
    Explanations

    references to specific programs or initiatives

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    +0.15
    0.9%
    356
    +0.13
    0.8%
    95
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    356
    +0.15
    0.03
    295
    +0.13
    0.03
    214
    +0.11
    0.02
    Negative Logits
    ĥ½
    -4.81
    ¥
    -4.09
    ¬
    -4.02
    ¶
    -3.93
    ®
    -3.74
    ·
    -3.69
    µ
    -3.66
    ĨĴ
    -3.65
    Ĵ
    -3.64
    IJ
    -3.62
    POSITIVE LOGITS
    matic
    2.75
    matically
    2.25
    atically
    2.17
    inator
    1.68
    inally
    1.63
    emaker
    1.57
     aimed
    1.57
     assistant
    1.56
     manager
    1.56
    etal
    1.55
    Activations Density 0.099%

    No Known Activations