© 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. 30311
    Prev
    Next
    INDEX
    Explanations

    instances of specific tokens or placeholders, likely representing variables, identifiers, or elements in programming or data structure 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
    410
    +0.14
    0.8%
    354
    +0.13
    0.7%
    350
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.14
    0.04
    230
    +0.13
    0.01
    271
    +0.12
    -0.01
    Negative Logits
     welcome
    -1.73
     hear
    -1.73
     battlefield
    -1.72
     face
    -1.58
     breeze
    -1.56
     look
    -1.56
     hereby
    -1.54
     forecast
    -1.53
     realize
    -1.48
     rehe
    -1.47
    POSITIVE LOGITS
    encial
    1.86
    plicity
    1.85
    ington
    1.71
    abase
    1.69
    antly
    1.67
    ious
    1.67
    atically
    1.64
    ĻĤ
    1.63
    endor
    1.62
    arro
    1.62
    Activations Density 0.415%

    No Known Activations