© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 12438
    Prev
    Next
    INDEX
    Explanations

    references to durations of time, specifically years and months

    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
    364
    +0.15
    0.8%
    258
    +0.10
    0.6%
    318
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.15
    0.06
    163
    +0.10
    0.04
    342
    +0.10
    0.04
    Negative Logits
    blogger
    -1.98
    acetic
    -1.61
    ery
    -1.57
     myself
    -1.44
    launchpad
    -1.41
     Blog
    -1.39
     forums
    -1.36
    CDATA
    -1.34
     proper
    -1.32
    phis
    -1.31
    POSITIVE LOGITS
     ago
    2.31
    chitz
    1.91
     elapsed
    1.87
     age
    1.84
    frame
    1.75
     expectancy
    1.75
    imar
    1.67
     old
    1.58
    OLD
    1.55
    cki
    1.53
    Activations Density 0.180%

    No Known Activations