© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 12377
    Prev
    Next
    INDEX
    Explanations

    instructions and parameters related to document editing and formatting features

    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
    326
    +0.18
    1.0%
    302
    +0.16
    0.9%
    419
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.18
    0.14
    401
    +0.16
    0.11
    412
    +0.16
    0.16
    Negative Logits
     Artists
    -1.69
     Jury
    -1.52
    gradle
    -1.49
     Cinema
    -1.40
     Ãħ
    -1.39
    ames
    -1.38
     Film
    -1.36
     Photoshop
    -1.31
    ANOVA
    -1.30
    Ļª
    -1.29
    POSITIVE LOGITS
     trouble
    1.70
    ailable
    1.57
     concerned
    1.56
     intended
    1.51
     awaited
    1.48
    '));
    1.46
    gered
    1.38
     worth
    1.37
     missed
    1.32
     needed
    1.30
    Activations Density 7.277%

    No Known Activations