© 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. 10079
    Prev
    Next
    INDEX
    Explanations

    references to time, specifically durations or time-related phrases

    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
    369
    +0.18
    1.0%
    46
    +0.14
    0.8%
    307
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.18
    0.03
    445
    +0.14
    0.02
    46
    +0.13
    0.02
    Negative Logits
     mistake
    -1.80
     history
    -1.78
     profits
    -1.70
     behalf
    -1.61
     forma
    -1.59
     possession
    -1.57
     responsibility
    -1.54
     destiny
    -1.53
     promise
    -1.48
     great
    -1.48
    POSITIVE LOGITS
    00
    2.05
    tees
    1.97
    teenth
    1.93
    ³
    1.91
    itrile
    1.87
    ĥ½
    1.74
    ´
    1.66
     inches
    1.63
    uncture
    1.60
    ®
    1.58
    Activations Density 0.173%

    No Known Activations