© 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. 4-RES-SM
    11. 13167
    Prev
    Next
    INDEX
    Explanations

    phrases or terms related to "proposals" or "propositions."

    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
    70
    +0.16
    0.9%
    203
    +0.14
    0.8%
    78
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.16
    0.04
    203
    +0.14
    0.03
    256
    +0.10
    0.03
    Negative Logits
    ering
    -1.89
    iginally
    -1.73
    ingle
    -1.61
     weeds
    -1.59
     contrasts
    -1.57
     unders
    -1.50
    assing
    -1.48
    rass
    -1.46
    iness
    -1.46
     woods
    -1.45
    POSITIVE LOGITS
    Ħ
    2.64
    ¾
    2.57
    ½
    2.46
    Ľ
    2.43
    blems
    2.28
    ducers
    2.24
    posal
    2.23
    Ŀ
    2.22
    ħ
    2.22
    ceedings
    2.19
    Activations Density 0.078%

    No Known Activations