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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 12068
    Prev
    Next
    INDEX
    Explanations

    terms related to the regulation and implications of free trade and its effects on various sectors

    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
    17
    +0.27
    1.5%
    186
    +0.14
    0.8%
    263
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.27
    0.44
    156
    +0.14
    0.40
    71
    +0.13
    0.38
    Negative Logits
    ¦
    -2.97
    ĥ
    -2.82
    ¿½
    -2.80
    Ŀ
    -2.78
    ¯
    -2.77
    ĸ
    -2.69
    ¨
    -2.69
    º
    -2.68
    Īĺ
    -2.68
                                                          
    -2.68
    POSITIVE LOGITS
     capacity
    1.69
     access
    1.58
     availability
    1.56
    ness
    1.54
     storage
    1.53
     rules
    1.52
     reproducibility
    1.51
     rule
    1.49
    capacity
    1.48
     applicability
    1.47
    Activations Density 2.213%

    No Known Activations