© 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. 24646
    Prev
    Next
    INDEX
    Explanations

    expressions of hope or expectation within the text

    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
    259
    +0.09
    0.5%
    469
    +0.09
    0.5%
    99
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.09
    0.02
    259
    +0.09
    0.02
    381
    +0.09
    0.01
    Negative Logits
     respectively
    -1.87
    thems
    -1.58
    TH
    -1.49
    ement
    -1.45
    Wil
    -1.41
    ruck
    -1.39
    CK
    -1.38
     depending
    -1.36
     stretching
    -1.34
    ening
    -1.34
    POSITIVE LOGITS
    ftware
    1.67
    amily
    1.60
    adjoint
    1.41
    printStackTrace
    1.41
    ility
    1.41
    upgreek
    1.38
    awt
    1.37
    otle
    1.33
    û
    1.32
    ffield
    1.32
    Activations Density 0.054%

    No Known Activations