© 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. 1-RES-SM
    11. 24858
    Prev
    Next
    INDEX
    Explanations

    references to systems or structured concepts in various 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    229
    +0.14
    0.8%
    223
    +0.11
    0.6%
    132
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.14
    -0.01
    23
    +0.11
    0.07
    88
    +0.11
    0.04
    Negative Logits
     contender
    -1.51
     album
    -1.46
    grass
    -1.42
     impression
    -1.39
     Award
    -1.39
     conservative
    -1.37
     spot
    -1.36
    azine
    -1.35
    weight
    -1.34
     heel
    -1.33
    POSITIVE LOGITS
    ´
    4.51
    Īĺ
    4.48
    Ļ
    4.13
    ĸ´
    4.09
    °
    4.08
    «
    4.08
    Ħ
    4.07
    ģ
    4.03
    Ī
    4.03
    Ľ
    3.99
    Activations Density 1.283%

    No Known Activations