© 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. 4-RES-SM
    11. 10328
    Prev
    Next
    INDEX
    Explanations

    phrases indicating time, especially those related to past and future events

    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
    156
    +0.15
    0.9%
    21
    +0.15
    0.9%
    400
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    291
    +0.15
    0.04
    169
    +0.15
    0.04
    250
    +0.13
    0.05
    Negative Logits
    ģ
    -3.03
    Ī
    -2.79
    ·¸
    -2.71
    ħ
    -2.69
    ī
    -2.69
    Ń
    -2.68
    ĥ
    -2.59
    Ģ
    -2.55
    º
    -2.53
    ±
    -2.51
    POSITIVE LOGITS
     entirety
    2.69
     years
    2.08
     entire
    1.97
     weekend
    1.92
     millenn
    1.91
     generations
    1.83
     centuries
    1.78
     next
    1.75
     whole
    1.75
     lifespan
    1.74
    Activations Density 0.292%

    No Known Activations