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. 11952
    Prev
    Next
    INDEX
    Explanations

    elements related to preparation and planning

    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
    234
    +0.12
    0.7%
    127
    +0.11
    0.6%
    190
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.12
    -0.02
    234
    +0.11
    0.12
    343
    +0.11
    0.10
    Negative Logits
    ´
    -7.50
    ¬
    -7.26
    ¿
    -7.22
    ²
    -7.21
    ħ
    -7.16
    Ħ
    -7.15
    ĨĴ
    -7.07
    ª
    -7.06
    Ń
    -7.00
    µ
    -6.99
    POSITIVE LOGITS
    "}](#
    1.62
    unless
    1.51
    ebra
    1.44
    fecture
    1.41
    picture
    1.39
     comparing
    1.36
     nowadays
    1.33
    PED
    1.33
    pgen
    1.31
    validate
    1.30
    Activations Density 2.912%

    No Known Activations