© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 14352
    Prev
    Next
    INDEX
    Explanations

    terms related to cloud computing technologies and platforms

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.23
    1.3%
    113
    +0.13
    0.7%
    444
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    113
    +0.23
    0.03
    304
    +0.13
    0.02
    444
    +0.12
    0.01
    Negative Logits
     miscar
    -1.75
    TRODUCTION
    -1.74
     vomiting
    -1.63
    plasia
    -1.61
    rodu
    -1.61
    nae
    -1.58
     hurry
    -1.57
    uppose
    -1.56
     menstrual
    -1.56
     vain
    -1.53
    POSITIVE LOGITS
    flare
    2.49
    ª
    2.16
    ľ
    2.14
    Ĵ
    2.01
    °
    1.96
    hart
    1.91
    ħ
    1.90
     computing
    1.89
    Ī
    1.79
    ķ
    1.77
    Activations Density 0.177%

    No Known Activations