© 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. 12405
    Prev
    Next
    INDEX
    Explanations

    references to various forms of work and employment concepts

    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
    184
    +0.15
    0.9%
    393
    +0.12
    0.7%
    318
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.15
    0.05
    318
    +0.12
    0.03
    375
    +0.12
    0.04
    Negative Logits
    Ĩ
    -2.91
    ĨĴ
    -2.55
    ľ
    -2.46
    ļ
    -2.40
    Ħ
    -2.39
    Ń
    -2.38
    IJ
    -2.37
    ij
    -2.27
    »
    -2.27
    ¯
    -2.26
    POSITIVE LOGITS
    station
    2.77
    manship
    2.62
    bench
    2.61
    places
    2.50
    horse
    2.48
    outs
    2.29
    room
    2.27
    place
    2.18
    book
    2.18
    books
    2.14
    Activations Density 0.163%

    No Known Activations