© 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. 12548
    Prev
    Next
    INDEX
    Explanations

    email addresses and data formats related to contact 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
    271
    +0.15
    0.8%
    353
    +0.11
    0.6%
    403
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.15
    0.00
    370
    +0.11
    0.06
    167
    +0.11
    0.03
    Negative Logits
    ?",
    -1.54
     {}
    -1.50
     peaks
    -1.46
     against
    -1.45
    itting
    -1.42
    ensity
    -1.42
    tons
    -1.41
     versus
    -1.41
     vary
    -1.40
     behaviours
    -1.34
    POSITIVE LOGITS
    ĻĤ
    2.11
    İ
    1.96
    IJ
    1.92
    ĭ
    1.91
    º
    1.89
    Ĩ
    1.87
    «
    1.86
    ģ
    1.86
    ³
    1.85
    ½
    1.82
    Activations Density 0.722%

    No Known Activations