© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 28463
    Prev
    Next
    INDEX
    Explanations

    instructions related to selection and filtering operations

    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
    287
    +0.17
    1.0%
    156
    +0.14
    0.8%
    339
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.17
    0.03
    339
    +0.14
    0.02
    148
    +0.12
    0.01
    Negative Logits
    ĥ½
    -2.92
    ĻĤ
    -2.62
    ĵ
    -2.50
    ³
    -2.33
    ľĵ
    -2.31
    ľ
    -2.26
    Ĭ
    -2.24
    ¿½
    -2.19
    į
    -2.16
    IJ
    -2.12
    POSITIVE LOGITS
    able
    2.07
    ably
    2.00
    ively
    1.62
     among
    1.60
    eting
    1.59
    legged
    1.57
    ables
    1.56
    ivity
    1.52
    pocket
    1.49
    aters
    1.49
    Activations Density 0.062%

    No Known Activations