© 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. 4-RES-SM
    11. 11347
    Prev
    Next
    INDEX
    Explanations

    elements related to specific names and titles

    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
    410
    +0.15
    0.8%
    274
    +0.13
    0.7%
    121
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.15
    0.03
    169
    +0.13
    0.05
    410
    +0.11
    0.05
    Negative Logits
    ifier
    -1.69
     heck
    -1.68
     Ubuntu
    -1.63
     user
    -1.61
     Android
    -1.59
     smartphone
    -1.57
     coronavirus
    -1.55
     teammate
    -1.54
     usability
    -1.54
     attacker
    -1.53
    POSITIVE LOGITS
    Illustration
    2.34
    Bibliography
    2.21
     Illustr
    2.00
    CHAPTER
    1.93
     LETTER
    1.92
    Ibid
    1.87
     poems
    1.85
     portraits
    1.84
    ]_
    1.84
     THE
    1.83
    Activations Density 1.274%

    No Known Activations