© 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. 1-RES-SM
    11. 27970
    Prev
    Next
    INDEX
    Explanations

    references to specific locations or names, particularly related to the term "Rock."

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    410
    +0.16
    0.9%
    477
    +0.13
    0.7%
    339
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.16
    0.02
    477
    +0.13
    0.02
    33
    +0.10
    0.02
    Negative Logits
     pregnancy
    -1.78
    usterity
    -1.71
     transl
    -1.58
     previous
    -1.54
     autom
    -1.51
     nitrogen
    -1.49
     initial
    -1.49
     particular
    -1.48
     chemo
    -1.48
     stress
    -1.47
    POSITIVE LOGITS
    ford
    2.34
    ings
    2.33
    ways
    2.27
    way
    2.19
    away
    2.17
    stead
    2.16
    port
    2.14
    aways
    2.08
    lane
    1.92
    yard
    1.91
    Activations Density 0.099%

    No Known Activations