© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 26674
    Prev
    Next
    INDEX
    Explanations

    recurring patterns or references to entities in a textual 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
    247
    +0.15
    0.8%
    444
    +0.11
    0.6%
    494
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    247
    +0.15
    0.03
    111
    +0.11
    0.03
    234
    +0.11
    0.03
    Negative Logits
    Ĭ
    -2.63
    ĵ
    -2.56
    Ł
    -2.54
    ¿½
    -2.47
    ¦
    -2.44
    Į
    -2.41
    Ĺ
    -2.40
    ĸ
    -2.37
    ¾
    -2.37
    ĻĤ
    -2.31
    POSITIVE LOGITS
    idemargin
    2.45
    pace
    2.45
    piration
    2.12
    mith
    2.06
    creen
    2.06
    chaft
    1.98
    àµį
    1.97
    ive
    1.95
    heet
    1.92
    ource
    1.88
    Activations Density 0.219%

    No Known Activations