© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 10770
    Prev
    Next
    INDEX
    Explanations

    phrases referring to events and time periods

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    56
    +0.15
    0.9%
    286
    +0.14
    0.8%
    190
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    190
    +0.15
    0.13
    286
    +0.14
    0.09
    68
    +0.14
    0.09
    Negative Logits
    Ģ
    -1.60
    ¿½
    -1.53
    tera
    -1.52
    ness
    -1.51
    anova
    -1.39
    ·¸
    -1.38
    ?",
    -1.34
    enance
    -1.34
    keit
    -1.33
     cytokine
    -1.31
    POSITIVE LOGITS
    oxford
    1.59
    StackTrace
    1.58
    aters
    1.37
    bered
    1.36
     Archae
    1.33
     Street
    1.30
    bid
    1.29
    isa
    1.28
    soon
    1.28
    wikipedia
    1.28
    Activations Density 3.176%

    No Known Activations