© 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. 29792
    Prev
    Next
    INDEX
    Explanations

    phrases related to time duration and measurement

    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
    263
    +0.21
    1.2%
    369
    +0.14
    0.8%
    291
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    291
    +0.21
    0.08
    263
    +0.14
    0.08
    436
    +0.13
    0.05
    Negative Logits
    £
    -3.14
    ĥ½
    -2.93
    ĥ
    -2.89
    ¢
    -2.85
    Īĺ
    -2.84
    ģ
    -2.84
    ¿½
    -2.79
    §
    -2.73
    ĩ
    -2.72
    ¹
    -2.71
    POSITIVE LOGITS
    ague
    1.43
     expiration
    1.43
     injection
    1.39
     completion
    1.38
     rounds
    1.37
    gio
    1.37
    dding
    1.35
    cember
    1.34
     depletion
    1.34
    ignon
    1.33
    Activations Density 0.983%

    No Known Activations