© 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. 31915
    Prev
    Next
    INDEX
    Explanations

    references to familial relationships and connections

    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
    302
    +0.24
    1.4%
    326
    +0.19
    1.1%
    233
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    302
    +0.24
    0.14
    326
    +0.19
    0.09
    223
    +0.13
    0.07
    Negative Logits
    woke
    -1.54
     destination
    -1.53
     mornings
    -1.53
     tempo
    -1.48
    aper
    -1.47
    upon
    -1.45
    Īĺ
    -1.44
    ľĵ
    -1.39
    inees
    -1.39
    ween
    -1.36
    POSITIVE LOGITS
    ção
    1.76
    uda
    1.63
    liness
    1.52
    ström
    1.45
     pleading
    1.41
    itism
    1.41
    AML
    1.39
    abel
    1.38
    onald
    1.35
    gal
    1.35
    Activations Density 5.955%

    No Known Activations