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. 12097
    Prev
    Next
    INDEX
    Explanations

    instances of numerical values and related statistical data

    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
    53
    +0.13
    0.8%
    206
    +0.12
    0.7%
    478
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.13
    0.08
    53
    +0.12
    0.10
    206
    +0.11
    0.05
    Negative Logits
    "}](#
    -2.04
    ural
    -1.56
     equivalents
    -1.56
    igner
    -1.47
    binding
    -1.47
    ourier
    -1.40
    core
    -1.40
    )|$(
    -1.39
     counterpart
    -1.38
    applicable
    -1.38
    POSITIVE LOGITS
    Ī
    2.83
    Į
    2.71
    ½
    2.59
    »
    2.52
    Ĵ
    2.51
    ¿½
    2.48
    ¼
    2.42
    ¢
    2.40
    Ļ
    2.39
    Ľ
    2.37
    Activations Density 0.745%

    No Known Activations