© 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. 13433
    Prev
    Next
    INDEX
    Explanations

    phrases expressing uncertainty or a need for assistance

    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
    416
    +0.15
    0.8%
    167
    +0.11
    0.6%
    391
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    91
    +0.15
    0.06
    492
    +0.11
    0.04
    243
    +0.11
    0.07
    Negative Logits
    tees
    -1.65
     representatives
    -1.51
    «
    -1.45
     Trophy
    -1.44
     evenly
    -1.39
    ionate
    -1.38
     Hawai
    -1.38
     reflects
    -1.36
    idas
    -1.33
    aterally
    -1.33
    POSITIVE LOGITS
     :-)
    2.08
     :)
    2.04
    .",
    1.98
     ;)
    1.86
    >>>
    1.85
    .</
    1.81
    :</
    1.76
    >\
    1.69
    ".
    1.68
    ?",
    1.68
    Activations Density 0.668%

    No Known Activations