© 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. 25257
    Prev
    Next
    INDEX
    Explanations

    phrases related to direction and movement

    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
    244
    +0.11
    0.6%
    368
    +0.11
    0.6%
    150
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    83
    +0.11
    0.04
    78
    +0.11
    0.04
    387
    +0.10
    0.04
    Negative Logits
    ĨĴ
    -2.46
    ŀ
    -2.38
    ®
    -2.21
    Ń
    -2.18
    İ
    -2.17
    Ŀ
    -2.08
    ł
    -2.07
    »
    -1.97
    ½
    -1.96
    ľ
    -1.94
    POSITIVE LOGITS
    agers
    1.64
    uries
    1.48
    onal
    1.47
     manual
    1.44
     controls
    1.43
     road
    1.43
     repair
    1.37
     quadr
    1.37
    ocamp
    1.36
     nuisance
    1.36
    Activations Density 0.551%

    No Known Activations