© 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 EvalsExportsinterp-engineNEWAPI 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. 12849
    Prev
    Next
    INDEX
    Explanations

    legal terminologies and processes related to court cases

    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
    156
    +0.67
    5.0%
    111
    +0.15
    1.1%
    240
    +0.14
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.67
    0.38
    71
    +0.15
    0.36
    111
    +0.14
    0.33
    Negative Logits
    IRST
    -1.36
    ÑĢоÑģ
    -1.29
    cco
    -1.28
    ftware
    -1.27
    ureus
    -1.25
    1451450014514500
    -1.23
    CLUDE
    -1.21
    tgz
    -1.20
    uego
    -1.16
    ó
    -1.12
    POSITIVE LOGITS
    Ģ
    2.04
    ļ
    1.80
    ĻĤ
    1.77
    Ĩ
    1.73
    Ħ
    1.73
    Į
    1.71
    ©
    1.70
    ī
    1.69
    į
    1.68
    ģ
    1.62
    Activations Density 5.736%

    No Known Activations