© 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. 9894
    Prev
    Next
    INDEX
    Explanations

    commands or prompts for user actions, particularly related to adding items or features

    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
    307
    +0.18
    1.0%
    240
    +0.14
    0.8%
    348
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.18
    0.03
    181
    +0.14
    0.04
    36
    +0.12
    0.04
    Negative Logits
    thing
    -1.56
     infra
    -1.37
    istically
    -1.35
    ually
    -1.35
     commune
    -1.33
    unnumbered
    -1.32
     BASIS
    -1.31
     Buff
    -1.30
    olysis
    -1.30
     sickness
    -1.30
    POSITIVE LOGITS
    itive
    3.12
    icted
    3.08
    Į
    3.07
    ¼
    3.04
    ressed
    2.94
    iction
    2.86
    §
    2.84
    ictions
    2.80
    ľĵ
    2.78
    Ħ
    2.78
    Activations Density 0.285%

    No Known Activations