© 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. 11362
    Prev
    Next
    INDEX
    Explanations

    key elements related to presenting research articles and findings

    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
    +0.31
    1.8%
    306
    +0.20
    1.2%
    466
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.31
    0.03
    418
    +0.20
    0.02
    307
    +0.12
    0.04
    Negative Logits
    acea
    -1.70
    ^âĪĴ
    -1.66
    inator
    -1.58
    bol
    -1.52
    ricular
    -1.51
    isible
    -1.47
    liness
    -1.43
    neys
    -1.43
    osphere
    -1.42
    eous
    -1.42
    POSITIVE LOGITS
     how
    2.26
     examples
    1.79
     terms
    1.75
     facts
    1.73
     exemplary
    1.70
    cases
    1.70
     descriptive
    1.65
     embodiments
    1.62
     salient
    1.59
     briefly
    1.58
    Activations Density 0.204%

    No Known Activations