© 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. 12557
    Prev
    Next
    INDEX
    Explanations

    phrases focused on personal involvement and responsibility

    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
    17
    +0.18
    1.0%
    320
    +0.16
    0.9%
    161
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    320
    +0.18
    0.08
    3
    +0.16
    0.07
    17
    +0.14
    0.02
    Negative Logits
    ī
    -4.13
    ŀ
    -4.08
    Ĭ
    -3.91
    ĸ
    -3.87
    Į
    -3.78
    ©
    -3.77
    Ļ
    -3.74
    ĵ
    -3.65
    ´
    -3.60
    -3.60
    POSITIVE LOGITS
    prising
    1.59
    est
    1.54
    eered
    1.51
    ient
    1.49
    eeper
    1.49
    ial
    1.49
     heard
    1.47
    plet
    1.45
    same
    1.40
    gered
    1.39
    Activations Density 0.805%

    No Known Activations