© 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. 29967
    Prev
    Next
    INDEX
    Explanations

    references to the name "Lin" and its variations

    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
    369
    +0.15
    0.8%
    156
    +0.14
    0.8%
    59
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.15
    0.05
    59
    +0.14
    0.03
    369
    +0.13
    0.02
    Negative Logits
     same
    -1.48
     sudden
    -1.47
     slightest
    -1.45
     host
    -1.42
     devil
    -1.39
     killers
    -1.38
    eeper
    -1.38
     younger
    -1.38
    bject
    -1.36
     utmost
    -1.32
    POSITIVE LOGITS
    emann
    2.21
    estone
    2.19
    quist
    1.97
    estock
    1.95
    acre
    1.94
    ª
    1.88
    hart
    1.82
    ¢
    1.74
    pton
    1.70
    owitz
    1.70
    Activations Density 0.254%

    No Known Activations