© 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. 14366
    Prev
    Next
    INDEX
    Explanations

    discourse markers signaling transitions or contrasts in discussions

    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
    307
    +0.29
    1.8%
    25
    +0.14
    0.8%
    58
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    476
    +0.29
    0.05
    307
    +0.14
    0.05
    330
    +0.13
    0.05
    Negative Logits
    ģ
    -3.11
    Ļª
    -2.88
    ħ
    -2.80
    ¿½
    -2.79
    Ħ
    -2.77
    ĥ
    -2.76
    ĵ
    -2.69
    ĨĴ
    -2.69
    ĸ
    -2.61
    Ĥ
    -2.59
    POSITIVE LOGITS
     my
    3.05
     I
    2.46
     mine
    2.18
    tons
    1.85
     when
    1.83
    terday
    1.83
     cmd
    1.82
     dropdown
    1.80
     btn
    1.75
     firefox
    1.74
    Activations Density 0.207%

    No Known Activations