© 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. 1-RES-SM
    11. 28774
    Prev
    Next
    INDEX
    Explanations

    Q values and message types in a programming context

    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
    230
    +0.14
    0.8%
    67
    +0.11
    0.6%
    164
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    67
    +0.14
    0.01
    230
    +0.11
    0.02
    232
    +0.10
    0.02
    Negative Logits
    than
    -1.63
     taxpayers
    -1.47
    Than
    -1.45
     entirety
    -1.44
    another
    -1.43
    ighs
    -1.43
    isecond
    -1.43
     called
    -1.41
    cules
    -1.40
    cted
    -1.39
    POSITIVE LOGITS
    naire
    1.85
     aloud
    1.59
    rams
    1.44
    abase
    1.44
    ullah
    1.43
    !(
    1.43
    !,
    1.43
    ERTY
    1.42
     assists
    1.41
    pril
    1.41
    Activations Density 0.105%

    No Known Activations