© 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. 13507
    Prev
    Next
    INDEX
    Explanations

    attributes related to technology and product features

    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
    419
    +0.20
    1.2%
    436
    +0.15
    0.9%
    271
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.20
    -0.01
    78
    +0.15
    0.05
    419
    +0.15
    0.07
    Negative Logits
    ľĵ
    -2.77
    ¯
    -2.01
    ¸
    -1.99
    ±
    -1.97
    §
    -1.94
    ª
    -1.92
    ĩ
    -1.82
    ¼
    -1.82
    Ļª
    -1.79
    Į
    -1.78
    POSITIVE LOGITS
    à±ģ
    1.80
    igraphy
    1.64
    à¯į
    1.55
     glance
    1.52
     discovery
    1.50
    à¯
    1.48
    ி
    1.47
    à°¿
    1.40
     briefs
    1.33
    à±
    1.33
    Activations Density 2.403%

    No Known Activations