© 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. 1-RES-SM
    11. 25517
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to the process of formation or creating something

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    376
    +0.12
    0.7%
    444
    +0.11
    0.6%
    93
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    444
    +0.12
    0.01
    93
    +0.11
    0.01
    470
    +0.11
    0.01
    Negative Logits
    ¤
    -1.66
    ĭ
    -1.58
    ĻĤ
    -1.57
    ¦
    -1.54
    ®
    -1.52
     ÑĩÑĤо
    -1.52
    =>
    -1.47
    Ľ
    -1.42
    wik
    -1.40
    Ĺ
    -1.39
    POSITIVE LOGITS
    idable
    1.86
    ulated
    1.83
    ulation
    1.69
    ères
    1.64
     them
    1.60
    ulating
    1.55
    osus
    1.54
    elic
    1.52
    ulate
    1.52
    atted
    1.50
    Activations Density 0.042%

    No Known Activations