© 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. 30998
    Prev
    Next
    INDEX
    Explanations

    words associated with various forms of corruption and cheating

    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
    261
    +0.13
    0.7%
    342
    +0.13
    0.7%
    419
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    13
    +0.13
    0.07
    21
    +0.13
    0.06
    261
    +0.11
    0.05
    Negative Logits
    blank
    -1.43
     someone
    -1.39
    ÅŁ
    -1.33
     retrieval
    -1.32
     Spacewatch
    -1.31
    odor
    -1.30
     year
    -1.29
    efore
    -1.29
     sooner
    -1.28
     courage
    -1.27
    POSITIVE LOGITS
    ftware
    2.48
    »¿
    1.92
    dal
    1.56
    dale
    1.52
    *.
    1.49
    boro
    1.48
    RY
    1.47
    ayer
    1.45
    Ĥ¬
    1.45
    .]
    1.41
    Activations Density 1.265%

    No Known Activations