© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 10665
    Prev
    Next
    INDEX
    Explanations

    instances of acknowledgment or recognition of information

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    25
    +0.17
    1.0%
    274
    +0.17
    0.9%
    58
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    472
    +0.17
    0.03
    485
    +0.17
    0.02
    330
    +0.13
    0.04
    Negative Logits
    ĥ
    -2.33
    Ŀ
    -2.19
    ľ
    -2.09
    Īĺ
    -2.00
    .’
    -1.98
    Ļª
    -1.98
    ģ
    -1.97
    ĵ
    -1.94
    .’”
    -1.94
    »¿
    -1.93
    POSITIVE LOGITS
     myself
    2.35
     workaround
    2.10
     tutorials
    2.02
     my
    1.98
     several
    1.97
     alot
    1.96
     some
    1.89
     checkbox
    1.79
     two
    1.78
     jquery
    1.74
    Activations Density 0.174%

    No Known Activations