© 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 EvalsExportsAPI 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. 27538
    Prev
    Next
    INDEX
    Explanations

    specific terms related to directives or commands, as well as code-related syntax

    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
    79
    +0.13
    0.7%
    274
    +0.12
    0.7%
    468
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    488
    +0.13
    0.02
    79
    +0.12
    0.01
    501
    +0.11
    0.01
    Negative Logits
    ¨
    -1.76
     nights
    -1.74
     survival
    -1.58
    ¾
    -1.57
    ª
    -1.52
     shred
    -1.51
     guts
    -1.49
     watching
    -1.48
     space
    -1.48
    °
    -1.47
    POSITIVE LOGITS
    ULAR
    1.83
    endent
    1.70
    ending
    1.66
    endant
    1.47
    иÑĢ
    1.43
    ê
    1.42
    nom
    1.39
    arm
    1.39
    along
    1.38
    ubern
    1.38
    Activations Density 0.102%

    No Known Activations