© 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. 30757
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to communication in various contexts

    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
    365
    +0.20
    1.2%
    451
    +0.16
    0.9%
    156
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.20
    0.03
    196
    +0.16
    0.01
    471
    +0.13
    0.01
    Negative Logits
    ½
    -1.99
     vengeance
    -1.87
     guilty
    -1.78
    ¼
    -1.75
    ·¸
    -1.69
    orem
    -1.66
    ľĵ
    -1.63
     revenge
    -1.58
    ®
    -1.55
    ²
    -1.54
    POSITIVE LOGITS
     channels
    2.02
     between
    1.89
     networks
    1.87
     centers
    1.85
     infrastructure
    1.82
     systems
    1.71
     devices
    1.71
    between
    1.70
     apparatus
    1.69
    iences
    1.65
    Activations Density 0.109%

    No Known Activations