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. 29503
    Prev
    Next
    INDEX
    Explanations

    mentions of specific operating systems and software-related terms

    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
    198
    +0.15
    0.9%
    346
    +0.13
    0.8%
    292
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.15
    0.08
    345
    +0.13
    0.07
    475
    +0.13
    0.05
    Negative Logits
     quo
    -1.84
     infertility
    -1.76
     death
    -1.53
    icable
    -1.47
     agenda
    -1.46
     warranted
    -1.42
     ambitious
    -1.41
     career
    -1.40
    DESC
    -1.32
     autoimmune
    -1.29
    POSITIVE LOGITS
    ooth
    1.68
    ]];
    1.67
    stic
    1.61
    Ń
    1.58
    ¢
    1.58
    .]{}
    1.57
    inger
    1.57
    .]
    1.51
    cdn
    1.51
    assen
    1.48
    Activations Density 1.278%

    No Known Activations