© 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. 11087
    Prev
    Next
    INDEX
    Explanations

    code related to functions and classes in programming

    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
    72
    +0.15
    0.9%
    302
    +0.11
    0.6%
    340
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.15
    0.10
    122
    +0.11
    0.09
    393
    +0.11
    0.07
    Negative Logits
     hell
    -1.57
    icable
    -1.39
    PEC
    -1.38
     fuck
    -1.36
    bras
    -1.34
    InstanceState
    -1.32
    mouth
    -1.30
     sake
    -1.28
     mattered
    -1.27
    krit
    -1.26
    POSITIVE LOGITS
    ¿½
    2.31
    ¿
    2.25
    ĻĤ
    2.06
    º
    2.01
    ¹
    1.98
    ħ
    1.96
    Ļ
    1.95
    ³
    1.94
    ))$.
    1.86
    ©
    1.84
    Activations Density 5.104%

    No Known Activations