© 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 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. 28855
    Prev
    Next
    INDEX
    Explanations

    terms relating to legal or policy-related language

    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
    12
    +0.18
    1.1%
    442
    +0.14
    0.8%
    198
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    12
    +0.18
    0.14
    78
    +0.14
    0.05
    455
    +0.13
    0.12
    Negative Logits
    ½
    -3.43
    ľĵ
    -3.27
    ĺ
    -3.25
    §
    -3.21
    ©
    -3.20
    ľ
    -3.06
    ¡
    -3.00
    ı
    -2.97
    ĨĴ
    -2.96
    ĻĤ
    -2.84
    POSITIVE LOGITS
     mattered
    1.77
    eries
    1.60
    ctrine
    1.54
    icable
    1.53
     issue
    1.49
    ricular
    1.47
    ifiable
    1.43
    asic
    1.42
    omy
    1.39
    etric
    1.38
    Activations Density 4.269%

    No Known Activations