© 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. 27841
    Prev
    Next
    INDEX
    Explanations

    financial terms and references related to customers and their 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
    153
    +0.20
    1.1%
    263
    +0.15
    0.9%
    204
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    153
    +0.20
    0.07
    204
    +0.15
    0.14
    335
    +0.13
    0.13
    Negative Logits
    ând
    -1.83
     Spacewatch
    -1.54
    sson
    -1.52
     hostage
    -1.42
    destruct
    -1.40
    asm
    -1.40
     Galaxy
    -1.40
    rely
    -1.35
    odend
    -1.33
    endo
    -1.32
    POSITIVE LOGITS
    ľĵ
    2.49
    ©
    2.43
    §
    2.28
    Į
    2.27
    ı
    2.21
    <|outofrange|>
    2.20
    2.20
    č↵
    2.20
    ↵↵
    2.20
    ↵
    2.20
    Activations Density 5.548%

    No Known Activations