© 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. 29403
    Prev
    Next
    INDEX
    Explanations

    technical jargon and specialized terminology, particularly from medical or scientific 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/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
    391
    +0.14
    0.8%
    135
    +0.12
    0.7%
    235
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    391
    +0.14
    0.03
    98
    +0.12
    -0.02
    452
    +0.12
    0.05
    Negative Logits
    Īĺ
    -2.41
    fulness
    -1.63
    aws
    -1.49
     tape
    -1.42
     basics
    -1.41
     neutrality
    -1.40
    acia
    -1.39
    gnu
    -1.39
     Federation
    -1.38
    ľĵ
    -1.36
    POSITIVE LOGITS
     että
    1.69
     Edition
    1.61
    yet
    1.59
    osti
    1.58
    amente
    1.53
    theless
    1.52
     solves
    1.50
    izumab
    1.45
    ño
    1.43
     unders
    1.40
    Activations Density 0.861%

    No Known Activations