© 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. 12668
    Prev
    Next
    INDEX
    Explanations

    terms related to high and low measurements in various contexts

    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
    341
    +0.12
    0.6%
    479
    +0.12
    0.6%
    11
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.12
    0.04
    341
    +0.12
    0.04
    11
    +0.10
    0.04
    Negative Logits
    osus
    -1.85
    soever
    -1.66
    UAL
    -1.62
    itors
    -1.55
    itor
    -1.55
    usiness
    -1.48
    ibles
    -1.47
    oths
    -1.45
    iting
    -1.44
     interviews
    -1.42
    POSITIVE LOGITS
    §
    2.72
    £
    2.65
    ģ
    2.62
    »
    2.61
    ļ
    2.59
    Īĺ
    2.52
    Ŀ
    2.51
    ĥ
    2.49
    ij
    2.46
    º
    2.46
    Activations Density 0.066%

    No Known Activations