© 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. 4-RES-SM
    11. 12842
    Prev
    Next
    INDEX
    Explanations

    negations and disclaimers regarding liability and limitations in legal documents

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    78
    +0.25
    1.4%
    105
    +0.13
    0.8%
    291
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.25
    0.04
    493
    +0.13
    0.04
    391
    +0.13
    0.04
    Negative Logits
    night
    -1.57
    ly
    -1.51
    ensively
    -1.47
    agne
    -1.41
    ression
    -1.40
    lies
    -1.40
    ieurs
    -1.39
     reception
    -1.35
    agner
    -1.31
    ypse
    -1.29
    POSITIVE LOGITS
    ħ
    4.49
    ĩ
    4.41
    Ĩ
    4.21
    İ
    4.15
    ĥ
    4.15
    Į
    4.14
    ģ
    4.10
    §
    4.09
    Ĥ
    4.05
    ĸ
    4.03
    Activations Density 0.128%

    No Known Activations