© 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. 28696
    Prev
    Next
    INDEX
    Explanations

    terms related to mathematical concepts and equations, particularly in physics and statistics

    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
    156
    +0.16
    0.9%
    456
    +0.12
    0.7%
    466
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.16
    0.04
    363
    +0.12
    0.03
    443
    +0.12
    0.02
    Negative Logits
     concerned
    -1.55
    hering
    -1.53
    ayers
    -1.48
    ourcing
    -1.48
    aring
    -1.48
    upon
    -1.45
    amed
    -1.42
    aked
    -1.41
    aming
    -1.39
    going
    -1.38
    POSITIVE LOGITS
    lite
    1.77
    keit
    1.76
    ismus
    1.71
    ian
    1.65
    schild
    1.51
    genstein
    1.49
    Ĵ
    1.48
    hill
    1.44
    fficients
    1.42
    ophers
    1.41
    Activations Density 0.175%

    No Known Activations