© 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. 13072
    Prev
    Next
    INDEX
    Explanations

    references to keyboard and mouse configurations or actions

    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
    344
    +0.14
    0.8%
    42
    +0.12
    0.6%
    498
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.14
    0.05
    225
    +0.12
    0.07
    154
    +0.11
    0.06
    Negative Logits
    ģ
    -2.14
    ¾
    -2.06
    ļ
    -1.97
    ĵ
    -1.91
    °
    -1.88
     SFR
    -1.83
    ħ
    -1.78
     Bradford
    -1.77
    Ł
    -1.77
    ³
    -1.75
    POSITIVE LOGITS
     keyboard
    3.82
     Ctrl
    3.38
     keyboards
    3.37
     cursor
    3.25
     ctrl
    2.95
     Mouse
    2.81
     mouse
    2.67
    pad
    2.53
    mouse
    2.45
     buttons
    2.42
    Activations Density 1.321%

    No Known Activations