© 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. 10125
    Prev
    Next
    INDEX
    Explanations

    function calls and method invocations

    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
    478
    +0.23
    1.3%
    86
    +0.15
    0.8%
    306
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.23
    0.05
    8
    +0.15
    0.04
    5
    +0.12
    0.05
    Negative Logits
    ĻĤ
    -2.28
    Ĵ
    -2.21
    ¿½
    -2.00
    ĩ
    -1.99
    ħ
    -1.91
    Ģ
    -1.90
    »¿
    -1.88
    ĸ
    -1.84
    ª
    -1.81
    ·
    -1.78
    POSITIVE LOGITS
     charset
    1.64
     cref
    1.37
    separated
    1.36
     dub
    1.34
    ↵↵
    1.34
     open
    1.31
     oligonucle
    1.29
    ↵↵
    1.29
     syscall
    1.26
     rounded
    1.25
    Activations Density 0.204%

    No Known Activations