© 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. 26593
    Prev
    Next
    INDEX
    Explanations

    articles and pronouns that indicate specificity

    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
    331
    +0.13
    0.7%
    138
    +0.12
    0.7%
    21
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.13
    0.07
    291
    +0.12
    0.05
    98
    +0.12
    -0.01
    Negative Logits
    _."
    -1.57
    .''
    -1.56
    .'"
    -1.55
    ."_
    -1.49
    stones
    -1.49
    __.
    -1.47
     burner
    -1.45
    .__
    -1.44
    uality
    -1.41
    dale
    -1.39
    POSITIVE LOGITS
    ermine
    1.53
     usual
    1.52
    verting
    1.50
    Willi
    1.49
    andum
    1.47
    lessly
    1.47
     ),
    1.42
     Sandy
    1.41
     slightest
    1.38
     famous
    1.36
    Activations Density 0.978%

    No Known Activations