© 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. 14011
    Prev
    Next
    INDEX
    Explanations

    snippets of code or technical programming structures

    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
    +0.18
    1.0%
    156
    +0.17
    1.0%
    231
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    491
    +0.18
    0.08
    271
    +0.17
    -0.01
    63
    +0.13
    0.06
    Negative Logits
    ).](
    -1.74
    >`
    -1.67
    ?’
    -1.64
    âĢħ
    -1.64
    \]](
    -1.62
    hores
    -1.55
    )](
    -1.48
    .[]{
    -1.47
    {};
    -1.40
    \-
    -1.38
    POSITIVE LOGITS
    ĻĤ
    3.20
    Ĥ
    3.18
    ¸
    3.09
    ¤
    3.06
    ¿
    3.01
    ¥
    2.97
    ĭ
    2.94
    º
    2.92
    ³
    2.91
    »
    2.91
    Activations Density 1.435%

    No Known Activations