© 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 EvalsExportsAPI 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. 27752
    Prev
    Next
    INDEX
    Explanations

    references to social media activities and metrics

    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
    156
    +0.17
    1.0%
    476
    +0.14
    0.8%
    79
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    476
    +0.17
    0.11
    95
    +0.14
    0.06
    79
    +0.11
    0.04
    Negative Logits
    ĸ´
    -5.56
    ĥ
    -5.44
    »¿
    -5.42
    ´
    -5.28
    ĥ½
    -5.22
    ↵
    -5.21
    č↵
    -5.21
    ↵
    -5.21
    č↵
    -5.21
    -5.21
    POSITIVE LOGITS
     websites
    2.10
     online
    2.08
     apps
    2.07
     hasht
    2.02
     website
    2.01
     downloads
    2.00
     URL
    2.00
    amazon
    1.97
     browser
    1.95
     internet
    1.95
    Activations Density 3.617%

    No Known Activations