© 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 EvalsExportsAPI 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. 10295
    Prev
    Next
    INDEX
    Explanations

    identifiers and keys associated with data structures in programming

    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
    156
    +0.33
    2.0%
    379
    +0.17
    1.0%
    17
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    379
    +0.33
    0.14
    17
    +0.17
    0.03
    129
    +0.14
    -0.00
    Negative Logits
     anion
    -1.45
    )\]
    -1.42
     suppose
    -1.41
    ureus
    -1.40
    ëĭĪëĭ¤
    -1.38
    ).](
    -1.38
    .)
    -1.38
     comparing
    -1.36
     ))
    -1.35
    ...)
    -1.35
    POSITIVE LOGITS
    _
    1.90
    manager
    1.56
    ahead
    1.53
     <-
    1.52
    ename
    1.50
    info
    1.45
    etable
    1.44
    ister
    1.42
    Ļª
    1.41
    illo
    1.36
    Activations Density 1.105%

    No Known Activations