© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29789
    Prev
    Next
    INDEX
    Explanations

    Java method calls and assertions related to queue operations

    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
    293
    +0.11
    0.6%
    12
    +0.11
    0.6%
    309
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.11
    0.15
    67
    +0.11
    0.11
    185
    +0.10
    0.11
    Negative Logits
    gado
    -1.47
     ones
    -1.46
     counterpart
    -1.44
     Les
    -1.32
     contrary
    -1.31
     origin
    -1.29
    .[]{
    -1.29
     exploration
    -1.28
     lately
    -1.25
     aim
    -1.24
    POSITIVE LOGITS
    ł
    4.08
    ij
    3.99
    ¿½
    3.98
    ¿
    3.88
    Ļª
    3.87
    ±
    3.75
    ¦
    3.73
    ·¸
    3.72
    Īĺ
    3.67
    ¾
    3.66
    Activations Density 1.457%

    No Known Activations