© 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. 9669
    Prev
    Next
    INDEX
    Explanations

    references to software products and their associated components or 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
    17
    +0.16
    1.0%
    169
    +0.14
    0.9%
    197
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    312
    +0.16
    0.11
    109
    +0.14
    0.07
    426
    +0.12
    0.08
    Negative Logits
    Ĥ¬
    -2.49
    ľĵ
    -2.29
    Ī
    -2.28
    ĨĴ
    -2.22
    ²
    -2.21
    ½
    -2.10
    §
    -2.06
    ĻĤ
    -2.06
    Į
    -2.02
    Ĺ
    -1.98
    POSITIVE LOGITS
     Encyclopedia
    1.80
     Suite
    1.72
     Catalog
    1.58
     Version
    1.55
     Edition
    1.54
     View
    1.53
     Files
    1.51
     Interface
    1.50
     dictionary
    1.47
     Format
    1.46
    Activations Density 3.409%

    No Known Activations