© 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. 13406
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial policies and their implications

    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
    +0.13
    0.7%
    400
    +0.12
    0.7%
    302
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.13
    0.12
    439
    +0.12
    0.06
    401
    +0.12
    0.08
    Negative Logits
    ĭ
    -3.10
    ħ
    -3.10
    į
    -3.08
    ¶
    -3.07
    ¹
    -3.02
    ¿½
    -2.96
    IJ
    -2.95
    Ļª
    -2.93
    ¨
    -2.93
    ĨĴ
    -2.89
    POSITIVE LOGITS
     distinction
    1.60
     elapsed
    1.50
    ETHOD
    1.49
     offence
    1.45
     positional
    1.42
     exception
    1.41
     value
    1.41
     drawbacks
    1.41
    esides
    1.40
     exceptions
    1.38
    Activations Density 6.003%

    No Known Activations