© 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. 11324
    Prev
    Next
    INDEX
    Explanations

    various forms of text and documentation, emphasizing their significance across different contexts

    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
    156
    +0.31
    1.8%
    466
    +0.14
    0.8%
    400
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    400
    +0.31
    0.03
    365
    +0.14
    0.06
    466
    +0.12
    0.05
    Negative Logits
    ]>
    -1.82
    ĸ´
    -1.73
     young
    -1.64
     boys
    -1.63
    Ĥ¬
    -1.58
     yours
    -1.57
     guys
    -1.54
     suppose
    -1.53
     another
    -1.52
    woke
    -1.50
    POSITIVE LOGITS
    icro
    1.73
    âĢIJ
    1.73
    worm
    1.67
    icum
    1.67
    ugin
    1.67
    istical
    1.65
    istry
    1.62
    ios
    1.58
    osition
    1.58
    casting
    1.56
    Activations Density 1.266%

    No Known Activations