© 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 EvalsExportsAPI 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. 9741
    Prev
    Next
    INDEX
    Explanations

    URLs and web links

    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
    231
    +0.15
    0.8%
    485
    +0.12
    0.7%
    35
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    485
    +0.15
    0.04
    425
    +0.12
    0.05
    221
    +0.12
    0.03
    Negative Logits
     extrinsic
    -1.67
     depend
    -1.61
     defective
    -1.55
     care
    -1.53
     differentiating
    -1.52
     quo
    -1.52
     engineering
    -1.51
     ascertain
    -1.48
     ambiguity
    -1.47
    termination
    -1.47
    POSITIVE LOGITS
    blogger
    2.77
    wordpress
    1.89
    blogspot
    1.84
     Blog
    1.81
     <%=
    1.75
    jee
    1.71
    ĨĴ
    1.66
    freak
    1.66
    noreply
    1.65
     celebrates
    1.61
    Activations Density 0.338%

    No Known Activations