© 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. 31629
    Prev
    Next
    INDEX
    Explanations

    numerical data related to demographic or sample characteristics

    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
    442
    +0.12
    0.7%
    358
    +0.11
    0.6%
    218
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.12
    0.02
    21
    +0.11
    0.06
    220
    +0.11
    0.04
    Negative Logits
    ¿½
    -3.12
    ¼
    -2.82
    ľ
    -2.81
    ¯
    -2.71
    ĻĤ
    -2.67
    ¨
    -2.60
    Ģ
    -2.57
    ¸
    -2.54
    º
    -2.54
    <|outofrange|>
    -2.54
    POSITIVE LOGITS
     owed
    1.58
     enrolled
    1.44
     age
    1.41
    rceil
    1.41
     capacity
    1.41
    ologie
    1.40
     hospital
    1.40
    atura
    1.38
    apse
    1.37
    cember
    1.37
    Activations Density 1.663%

    No Known Activations