© 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. 2-RES-SM
    11. 27532
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "per" indicating rates or ratios

    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
    11
    +0.15
    0.8%
    53
    +0.13
    0.7%
    155
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    11
    +0.15
    0.03
    501
    +0.13
    0.02
    155
    +0.12
    0.03
    Negative Logits
    ETHOD
    -1.73
    Inflater
    -1.57
    resses
    -1.57
    rapper
    -1.49
    ³³³³³³³³
    -1.48
    uben
    -1.48
    [_
    -1.46
    OUP
    -1.45
    \|_{
    -1.42
    owski
    -1.41
    POSITIVE LOGITS
    Ħ
    2.07
    ished
    2.06
    ¾
    2.01
     capita
    2.01
    vious
    1.98
    ¿
    1.97
     curiam
    1.88
    imeters
    1.87
     hour
    1.85
    Ļª
    1.85
    Activations Density 0.062%

    No Known Activations