© 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. 27393
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'w'

    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
    410
    +0.14
    0.8%
    77
    +0.13
    0.7%
    317
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.14
    0.03
    77
    +0.13
    0.03
    317
    +0.11
    0.03
    Negative Logits
    igraph
    -1.53
    ancer
    -1.52
    estion
    -1.47
    course
    -1.40
    eered
    -1.35
    unity
    -1.35
    rance
    -1.33
     maps
    -1.33
    cient
    -1.33
    Ĥ
    -1.32
    POSITIVE LOGITS
    elcome
    1.81
    avy
    1.77
    avelength
    1.70
    olver
    1.66
    nuts
    1.64
    arts
    1.63
    agner
    1.62
    aiver
    1.62
    grant
    1.60
    igs
    1.59
    Activations Density 0.120%

    No Known Activations