© 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. 2-RES-SM
    11. 12864
    Prev
    Next
    INDEX
    Explanations

    references to specific events or occurrences

    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
    320
    +0.14
    0.8%
    245
    +0.12
    0.7%
    178
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.14
    0.16
    156
    +0.12
    0.17
    71
    +0.11
    0.15
    Negative Logits
     certificates
    -1.48
    aid
    -1.46
    ceral
    -1.44
    ocur
    -1.41
    illery
    -1.41
    chains
    -1.38
    istry
    -1.37
    olutions
    -1.36
    erala
    -1.35
    CLUDING
    -1.35
    POSITIVE LOGITS
    ked
    1.50
     instant
    1.35
     reference
    1.32
     later
    1.32
     hope
    1.32
     {-
    1.29
     diz
    1.28
    ursor
    1.26
     terminate
    1.25
     him
    1.25
    Activations Density 2.560%

    No Known Activations