© 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. 28742
    Prev
    Next
    INDEX
    Explanations

    programming structures or components related to data handling or user properties

    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
    497
    +0.12
    0.7%
    287
    +0.12
    0.6%
    464
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.12
    0.03
    82
    +0.12
    0.05
    271
    +0.11
    -0.02
    Negative Logits
    ";
    -1.61
    :`
    -1.51
    Argued
    -1.43
     meant
    -1.43
    Invitrogen
    -1.43
    needed
    -1.40
    rugged
    -1.40
    $)
    -1.39
    SON
    -1.37
    $–
    -1.36
    POSITIVE LOGITS
    ĻĤ
    3.61
    »¿
    3.28
    Ļª
    3.24
    ·¸
    3.05
    ŀ
    3.00
    Ĵ
    2.97
    ĸ´
    2.96
    ¼
    2.94
    Ŀ
    2.90
    ¸
    2.89
    Activations Density 1.177%

    No Known Activations