© 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. 9729
    Prev
    Next
    INDEX
    Explanations

    references to numerical values and codes

    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
    203
    +0.15
    0.8%
    78
    +0.12
    0.7%
    196
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    458
    +0.15
    0.01
    136
    +0.12
    -0.01
    477
    +0.12
    0.03
    Negative Logits
     Pradesh
    -1.55
     possible
    -1.54
    iances
    -1.52
    ffect
    -1.47
    ches
    -1.44
     other
    -1.44
    htra
    -1.44
     respective
    -1.43
    ering
    -1.42
    ened
    -1.33
    POSITIVE LOGITS
    Ļ
    4.78
    Ł
    4.53
    į
    4.36
    Ľ
    4.20
    Ķ
    4.13
    ĩ
    4.08
    Ĵ
    4.07
    ĵ
    4.04
    ½
    3.96
    »
    3.93
    Activations Density 0.196%

    No Known Activations