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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 30962
    Prev
    Next
    INDEX
    Explanations

    terms related to programming and data handling

    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
    47
    +0.15
    0.8%
    157
    +0.14
    0.8%
    379
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.15
    0.11
    271
    +0.14
    -0.00
    313
    +0.12
    0.06
    Negative Logits
    dens
    -1.63
    .).
    -1.44
    than
    -1.41
    .);
    -1.37
    den
    -1.37
    unreadable
    -1.37
    .]{}
    -1.34
    ]{}
    -1.34
     Stre
    -1.33
    Lyn
    -1.32
    POSITIVE LOGITS
    еÐ
    1.78
    Enabled
    1.76
    ption
    1.71
     <-
    1.71
    _
    1.66
    Policy
    1.66
    @
    1.65
    bserver
    1.65
    ename
    1.64
    picker
    1.62
    Activations Density 1.290%

    No Known Activations