© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11704
    Prev
    Next
    INDEX
    Explanations

    elements related to licensing and permission terms in software documentation

    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
    156
    +0.27
    1.6%
    21
    +0.16
    0.9%
    328
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    328
    +0.27
    0.04
    213
    +0.16
    0.06
    162
    +0.11
    0.02
    Negative Logits
    ↵
    -3.75
    ↵
    -3.75
    ↵
    -3.75
    ↵
    -3.75
    ↵
    -3.75
    č↵
    -3.75
    <|outofrange|>
    -3.75
    <|outofrange|>
    -3.75
    <|padding|>
    -3.75
    ↵
    -3.75
    POSITIVE LOGITS
    LICENSE
    3.06
     COPYRIGHT
    2.89
     Redistributions
    2.66
     LICENSE
    2.65
     copyright
    2.62
     License
    2.49
    license
    2.42
     Copyright
    2.42
    GPL
    2.30
    License
    2.26
    Activations Density 0.669%

    No Known Activations