© 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. 28884
    Prev
    Next
    INDEX
    Explanations

    methods and descriptions in procedural or instructional contexts

    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
    263
    +0.18
    1.0%
    198
    +0.14
    0.8%
    478
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.18
    0.09
    278
    +0.14
    0.06
    130
    +0.13
    0.11
    Negative Logits
    »¿
    -2.43
    į
    -2.20
    ¿½
    -2.06
    ľ
    -1.97
    «
    -1.90
    ¡
    -1.86
     possessions
    -1.81
    ¾
    -1.76
    ·
    -1.74
    ĭ
    -1.74
    POSITIVE LOGITS
    apest
    1.95
    method
    1.75
     methods
    1.64
     method
    1.59
     iterative
    1.59
     optimization
    1.57
    asone
    1.55
     inductive
    1.51
     techniques
    1.46
    designed
    1.43
    Activations Density 2.114%

    No Known Activations