© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 11159
    Prev
    Next
    INDEX
    Explanations

    references to specific years and legal citations

    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
    442
    +0.14
    0.8%
    271
    +0.14
    0.8%
    264
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    58
    +0.14
    0.04
    471
    +0.14
    0.03
    253
    +0.11
    0.03
    Negative Logits
    ions
    -1.66
     possibilities
    -1.63
    encies
    -1.63
     â̦
    -1.59
     affairs
    -1.58
     powers
    -1.53
    ities
    -1.52
     sizes
    -1.51
    iences
    -1.47
     waters
    -1.44
    POSITIVE LOGITS
    ĨĴ
    3.16
    Ļª
    2.79
    ±
    2.71
    »
    2.51
    ĻĤ
    2.50
    ¯
    2.48
    ĸ´
    2.42
    ŀ
    2.39
    °
    2.37
    ·
    2.36
    Activations Density 0.125%

    No Known Activations