© 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. 4-RES-SM
    11. 9206
    Prev
    Next
    INDEX
    Explanations

    references to specific categories or classifications

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    343
    +0.14
    0.8%
    437
    +0.11
    0.6%
    415
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    437
    +0.14
    0.03
    148
    +0.11
    0.02
    343
    +0.11
    0.02
    Negative Logits
     generality
    -1.79
    WARE
    -1.55
    ing
    -1.55
     recent
    -1.47
    houses
    -1.44
     comprehension
    -1.40
     subsequent
    -1.38
     Matter
    -1.38
     arsenal
    -1.34
    potent
    -1.31
    POSITIVE LOGITS
    ²
    4.07
    ĻĤ
    3.60
    ĥ
    3.48
    Īĺ
    3.44
    ľĵ
    3.34
    «
    3.32
    Ĥ
    3.23
    Į
    3.19
    ĸ
    3.19
    º
    3.19
    Activations Density 0.095%

    No Known Activations