© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 30223
    Prev
    Next
    INDEX
    Explanations

    programming concepts and structures related to functions and variables

    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
    186
    +0.15
    0.9%
    72
    +0.14
    0.8%
    122
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.15
    0.75
    156
    +0.14
    0.69
    71
    +0.11
    0.67
    Negative Logits
    ')->
    -1.74
    ERTY
    -1.59
     maiden
    -1.55
    )>
    -1.47
    arda
    -1.45
    msgstr
    -1.44
    $)
    -1.43
    &
    -1.43
    lit
    -1.39
     arrangements
    -1.38
    POSITIVE LOGITS
    ¿½
    1.77
    ¦
    1.37
    ŀ
    1.36
     Infantry
    1.36
    uously
    1.36
    ģ
    1.32
    rass
    1.32
    woke
    1.32
     Va
    1.31
    ACP
    1.29
    Activations Density 6.735%

    No Known Activations