© 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. 13986
    Prev
    Next
    INDEX
    Explanations

    phrases related to avoidance or evasion

    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
    115
    +0.13
    0.7%
    358
    +0.12
    0.7%
    465
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    465
    +0.13
    0.04
    358
    +0.12
    0.04
    327
    +0.12
    0.03
    Negative Logits
    ģ
    -1.98
    Ģ
    -1.94
    Ħ
    -1.87
    ĥ
    -1.77
    ±
    -1.74
    ¦
    -1.73
     comprised
    -1.71
    Ĺ
    -1.63
    ARY
    -1.57
    ħ
    -1.56
    POSITIVE LOGITS
    igne
    1.81
    omnia
    1.59
     temptation
    1.57
     collisions
    1.53
    erce
    1.53
     mistakes
    1.48
    uthan
    1.46
     evaporation
    1.45
    cessive
    1.45
    hib
    1.45
    Activations Density 0.300%

    No Known Activations