© 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. 26366
    Prev
    Next
    INDEX
    Explanations

    instances of the word "to" and its variations, indicating a focus on infinitive verbs and their uses

    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
    70
    +0.13
    0.7%
    75
    +0.12
    0.7%
    74
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.13
    0.08
    75
    +0.12
    0.03
    231
    +0.12
    0.06
    Negative Logits
    ↵
    -1.60
     NOTICE
    -1.45
     inspiration
    -1.44
     Description
    -1.44
     carcinogenesis
    -1.43
    ancers
    -1.42
    :**]{}
    -1.38
    acular
    -1.35
    *](#
    -1.35
    ankind
    -1.33
    POSITIVE LOGITS
    rij
    1.63
    rip
    1.62
    quit
    1.56
    gnu
    1.54
    parent
    1.52
     punish
    1.48
     divorce
    1.46
    option
    1.45
    kill
    1.44
    onies
    1.43
    Activations Density 1.292%

    No Known Activations