© 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. 9966
    Prev
    Next
    INDEX
    Explanations

    questions marked by a question mark symbol

    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
    21
    +0.31
    1.8%
    58
    +0.13
    0.7%
    103
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.31
    0.02
    96
    +0.13
    0.01
    455
    +0.12
    0.01
    Negative Logits
    Īĺ
    -2.94
    ¶
    -2.64
    ĸ
    -2.52
    ĥ
    -2.51
    Ĥ¬
    -2.44
    Ī
    -2.43
    Ħ
    -2.41
    £
    -2.41
    ¾
    -2.40
    ¿
    -2.35
    POSITIVE LOGITS
     Does
    2.17
     How
    1.93
     Why
    1.92
     Seems
    1.82
     Are
    1.80
     Probably
    1.79
     Then
    1.76
     Would
    1.76
     Isn
    1.74
     Something
    1.71
    Activations Density 0.036%

    No Known Activations