© 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. 9556
    Prev
    Next
    INDEX
    Explanations

    references or citations within academic texts

    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
    87
    +0.13
    0.7%
    478
    +0.13
    0.7%
    222
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    221
    +0.13
    0.02
    222
    +0.13
    0.02
    87
    +0.12
    0.01
    Negative Logits
     responding
    -1.70
    shire
    -1.63
    ESULT
    -1.57
    rage
    -1.45
    inating
    -1.40
    door
    -1.34
     polling
    -1.33
     permitting
    -1.32
    frastructure
    -1.31
    ellate
    -1.31
    POSITIVE LOGITS
    İ
    1.60
    aways
    1.58
     rib
    1.51
    ques
    1.41
    IJ
    1.39
     heart
    1.39
    ksi
    1.38
    fileID
    1.37
     transl
    1.36
     rays
    1.35
    Activations Density 0.137%

    No Known Activations