© 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. 9714
    Prev
    Next
    INDEX
    Explanations

    numerical values and their associated significance in a technical context

    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
    43
    +0.13
    0.8%
    461
    +0.13
    0.7%
    478
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.13
    0.04
    43
    +0.13
    0.03
    393
    +0.12
    0.03
    Negative Logits
    ]{}.
    -1.68
    *]{}.
    -1.59
    ester
    -1.54
    ]{},
    -1.47
     Wing
    -1.46
    ’.
    -1.39
    otrophic
    -1.38
     estimated
    -1.38
    ]:
    -1.38
    enger
    -1.37
    POSITIVE LOGITS
    ·¸
    3.47
    ¿½
    3.16
    ĥ½
    3.03
    Īĺ
    2.90
    Ĥ
    2.86
    ľĵ
    2.85
    ĭ
    2.74
    Ľ
    2.73
    ģ
    2.69
    ↵
    2.69
    Activations Density 0.272%

    No Known Activations