© 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. 4-RES-SM
    11. 10419
    Prev
    Next
    INDEX
    Explanations

    expressions of desire, preference, or intention

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    43
    +0.15
    0.8%
    315
    +0.11
    0.6%
    444
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    26
    +0.15
    0.10
    358
    +0.11
    0.10
    272
    +0.10
    0.06
    Negative Logits
    Ĵ
    -4.01
    Ģ
    -3.93
    Ļª
    -3.85
    Ħ
    -3.80
    ķ
    -3.79
    ĻĤ
    -3.77
    ĸ
    -3.76
    ŀ
    -3.72
    Ŀ
    -3.71
    ®
    -3.65
    POSITIVE LOGITS
     anymore
    3.78
     nor
    2.41
     nada
    2.35
     any
    2.25
     anything
    2.21
     unnecessarily
    2.18
     whatsoever
    2.16
     overly
    2.15
     Instead
    2.14
     anywhere
    2.12
    Activations Density 1.166%

    No Known Activations