© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26388
    Prev
    Next
    INDEX
    Explanations

    URLs and references to licensing or copyright information

    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
    446
    +0.12
    0.6%
    26
    +0.11
    0.6%
    222
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    315
    +0.12
    0.03
    74
    +0.11
    0.03
    222
    +0.11
    0.03
    Negative Logits
    thro
    -1.86
     however
    -1.70
    rait
    -1.51
     injuries
    -1.37
    PCs
    -1.35
    vez
    -1.32
    pite
    -1.32
    eter
    -1.31
    landers
    -1.30
    bral
    -1.29
    POSITIVE LOGITS
     adj
    1.66
     libr
    1.55
    ublic
    1.52
     author
    1.47
     publisher
    1.46
    ŀ
    1.43
     codon
    1.41
    bound
    1.39
     unpublished
    1.39
     kinase
    1.37
    Activations Density 0.253%

    No Known Activations