© 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. 10633
    Prev
    Next
    INDEX
    Explanations

    legal vernacular related to testimony and evidence in court proceedings

    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
    307
    +0.14
    0.8%
    47
    +0.12
    0.6%
    211
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.14
    0.08
    44
    +0.12
    0.09
    307
    +0.10
    0.11
    Negative Logits
     accordingly
    -1.76
    ronic
    -1.47
    nant
    -1.41
    rums
    -1.40
     ensued
    -1.37
    ega
    -1.32
    !_
    -1.31
    tico
    -1.31
     versions
    -1.30
     revised
    -1.30
    POSITIVE LOGITS
     someone
    2.29
     somebody
    1.96
     Someone
    1.93
     something
    1.91
    someone
    1.90
     an
    1.89
     a
    1.77
     oneself
    1.73
    something
    1.71
     certain
    1.61
    Activations Density 1.853%

    No Known Activations