© 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. 4-RES-SM
    11. 13555
    Prev
    Next
    INDEX
    Explanations

    components and attributes related to graphical figures and documents

    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
    78
    +0.17
    1.0%
    477
    +0.12
    0.7%
    9
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    9
    +0.17
    0.04
    325
    +0.12
    0.04
    78
    +0.12
    0.00
    Negative Logits
    oretic
    -1.64
     ‘
    -1.53
     now
    -1.47
     £
    -1.42
     internally
    -1.41
     spokesperson
    -1.39
    ching
    -1.39
     trained
    -1.38
     organisation
    -1.37
     cheapest
    -1.36
    POSITIVE LOGITS
    ĭ
    5.62
    Ĥ
    5.43
    Īĺ
    5.41
    ²
    5.36
    ĥ½
    5.34
             ↵
    5.34
    <|outofrange|>
    5.34
    <|outofrange|>
    5.34
    ↵↵
    5.34
    ↵↵
    5.34
    Activations Density 0.081%

    No Known Activations