© 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. 27162
    Prev
    Next
    INDEX
    Explanations

    structured data and metadata in a specific format

    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
    24
    +0.14
    0.8%
    436
    +0.14
    0.8%
    153
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    226
    +0.14
    0.10
    212
    +0.14
    0.12
    24
    +0.14
    0.13
    Negative Logits
    Īĺ
    -2.55
    º
    -2.39
    ¹
    -2.13
    ĸ
    -1.94
    »¿
    -1.91
    ¿½
    -1.86
    ¼
    -1.85
    ļ
    -1.82
    ¶
    -1.80
    Ĺ
    -1.79
    POSITIVE LOGITS
     includes
    1.32
    )_
    1.29
    ove
    1.27
     contains
    1.27
     Helsinki
    1.26
     omit
    1.23
     reference
    1.23
     Statistics
    1.22
     emphasis
    1.22
    emph
    1.22
    Activations Density 7.097%

    No Known Activations