© 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. 2-RES-SM
    11. 10564
    Prev
    Next
    INDEX
    Explanations

    code snippets related to data types and structures in programming

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    449
    +0.13
    0.7%
    273
    +0.12
    0.7%
    442
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    491
    +0.13
    0.07
    220
    +0.12
    0.06
    261
    +0.12
    0.05
    Negative Logits
    !'
    -1.74
    !’
    -1.59
    clamation
    -1.56
    Bot
    -1.46
    ots
    -1.43
     correctly
    -1.41
    !\
    -1.41
    .[]{
    -1.37
     syntax
    -1.34
    **:
    -1.33
    POSITIVE LOGITS
     etc
    1.62
    teenth
    1.50
    etc
    1.47
     otros
    1.46
     countryside
    1.42
    greg
    1.42
    iento
    1.41
     NSString
    1.40
     respectively
    1.38
    respectively
    1.37
    Activations Density 0.773%

    No Known Activations