© 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. 32041
    Prev
    Next
    INDEX
    Explanations

    terms related to plasma and its properties or measurements

    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
    156
    +0.22
    1.3%
    339
    +0.12
    0.7%
    164
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.22
    0.01
    148
    +0.12
    0.00
    45
    +0.10
    0.01
    Negative Logits
    ĥ½
    -2.56
    Ń
    -2.35
    ½
    -2.27
    §
    -2.18
    ¾
    -2.08
    į
    -2.05
    ³
    -2.05
    ¥
    -2.05
    Ķ
    -2.03
    ¶
    -1.98
    POSITIVE LOGITS
    ellar
    2.10
    rice
    1.85
     concentrations
    1.73
     levels
    1.64
    ster
    1.56
    illin
    1.52
     values
    1.51
    ime
    1.44
    sky
    1.44
     Values
    1.43
    Activations Density 0.037%

    No Known Activations