© 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. 31066
    Prev
    Next
    INDEX
    Explanations

    information related to graphical user interface elements or commands

    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
    282
    +0.12
    0.7%
    294
    +0.12
    0.7%
    257
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    374
    +0.12
    0.00
    421
    +0.12
    0.00
    228
    +0.11
    0.00
    Negative Logits
    ':
    -1.79
    ":
    -1.61
     outset
    -1.55
    bler
    -1.53
    edo
    -1.52
    escence
    -1.51
    ":{
    -1.50
    ”:
    -1.46
    assertEqual
    -1.45
    len
    -1.45
    POSITIVE LOGITS
     precautions
    1.65
     unclear
    1.54
     multivariate
    1.49
    ally
    1.45
     worsening
    1.39
    seeing
    1.39
     probability
    1.39
     professionally
    1.38
     cumulative
    1.34
    ively
    1.33
    Activations Density 0.000%

    No Known Activations