© 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. 1-RES-SM
    11. 24849
    Prev
    Next
    INDEX
    Explanations

    details about scientific classifications and comparisons between species

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    23
    +0.20
    1.1%
    271
    +0.16
    0.9%
    0
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.20
    0.28
    478
    +0.16
    -0.06
    152
    +0.11
    0.16
    Negative Logits
     Bachelor
    -1.74
    "];
    -1.55
    haus
    -1.52
     Minist
    -1.49
     humility
    -1.46
     manners
    -1.43
    hearted
    -1.42
     lessons
    -1.42
     WARRANTIES
    -1.41
    ensch
    -1.41
    POSITIVE LOGITS
    Errno
    1.97
    ```
    1.70
    esium
    1.59
    LEX
    1.46
    respectively
    1.43
     poll
    1.41
    ixel
    1.40
    ycler
    1.39
    icon
    1.38
    ipes
    1.37
    Activations Density 11.819%

    No Known Activations