© 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. 9044
    Prev
    Next
    INDEX
    Explanations

    specific numerical data or statistical information

    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
    477
    +0.13
    0.7%
    451
    +0.13
    0.7%
    21
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    477
    +0.13
    0.06
    458
    +0.13
    0.03
    129
    +0.13
    0.01
    Negative Logits
    ¿½
    -2.59
    ĭ
    -2.57
    Ĥ
    -2.56
    ↵↵
    -2.50
    ↵↵
    -2.50
    ↵
    -2.50
    -2.50
     ↵
    -2.50
          ↵
    -2.50
    ↵
    -2.50
    POSITIVE LOGITS
    04
    2.48
    05
    2.10
    06
    2.09
    07
    2.08
    aged
    2.07
    042
    2.03
    092
    2.01
    62
    2.00
    03
    1.98
    08
    1.97
    Activations Density 0.269%

    No Known Activations