© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29896
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'K'

    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
    +0.20
    1.2%
    156
    +0.14
    0.8%
    171
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.20
    0.03
    410
    +0.14
    0.06
    59
    +0.12
    0.03
    Negative Logits
    cluding
    -1.69
    ductive
    -1.69
    herence
    -1.67
    effective
    -1.47
    cluded
    -1.41
    cess
    -1.40
    uppose
    -1.40
    actory
    -1.38
    orters
    -1.37
     previous
    -1.37
    POSITIVE LOGITS
    ĨĴ
    2.04
    ĻĤ
    1.96
    ¿
    1.93
    rish
    1.89
    Į
    1.87
    Ĵ
    1.87
    inetics
    1.81
    ings
    1.77
    iewicz
    1.75
    ĸ
    1.72
    Activations Density 0.255%

    No Known Activations