Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 11790
    Prev
    Next
    INDEX
    Explanations

    technical terms and functionalities related to programming and file management

    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
    +0.36
    2.1%
    31
    +0.13
    0.8%
    95
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.36
    0.05
    400
    +0.13
    0.02
    240
    +0.11
    -0.01
    Negative Logits
    ĵ
    -2.03
    ĥ½
    -1.90
    ĥ
    -1.90
    ·¸
    -1.81
    Ĥ¬
    -1.73
    ij
    -1.62
    ĨĴ
    -1.56
     attendant
    -1.53
    Ŀ
    -1.52
    ĭ
    -1.52
    POSITIVE LOGITS
    code
    1.77
    bytes
    1.70
    .",
    1.66
    boolean
    1.60
     isn
    1.58
     '%
    1.57
    lookup
    1.53
    TeX
    1.53
     =>
    1.53
     integer
    1.52
    Activations Density 0.643%

    No Known Activations