© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12522
    Prev
    Next
    INDEX
    Explanations

    structured numerical data or identifiers

    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
    +0.30
    1.7%
    13
    +0.14
    0.8%
    153
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.30
    0.04
    13
    +0.14
    0.11
    79
    +0.13
    0.09
    Negative Logits
    ¶
    -2.85
    Ī
    -2.80
    ī
    -2.61
    ¦
    -2.60
    Į
    -2.57
    µ
    -2.56
    º
    -2.53
    Ŀ
    -2.52
    ĺ
    -2.50
    §
    -2.47
    POSITIVE LOGITS
    ftware
    1.43
     $@
    1.40
    fessional
    1.39
    igent
    1.35
    sburg
    1.34
     */,
    1.32
    ramento
    1.32
    oard
    1.31
    leans
    1.28
     Rivers
    1.28
    Activations Density 3.057%

    No Known Activations