© 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. 27106
    Prev
    Next
    INDEX
    Explanations

    corporation and company names

    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
    350
    +0.16
    0.9%
    410
    +0.15
    0.9%
    83
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.16
    0.05
    350
    +0.15
    0.04
    59
    +0.13
    0.02
    Negative Logits
     inspiration
    -1.72
     weddings
    -1.55
     contrary
    -1.51
     wisely
    -1.46
     inheritance
    -1.45
    rd
    -1.40
    ize
    -1.39
     attention
    -1.37
    ibernate
    -1.36
     obvious
    -1.33
    POSITIVE LOGITS
    ¿½
    2.62
    ¾
    2.12
    «
    2.04
    ²
    1.97
    ľ
    1.96
    ulence
    1.92
    atures
    1.78
    ¿
    1.78
    ships
    1.77
    uscular
    1.75
    Activations Density 0.333%

    No Known Activations