© 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. 14675
    Prev
    Next
    INDEX
    Explanations

    interrogative words asking for information or details

    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
    111
    +0.58
    4.9%
    156
    +0.55
    4.6%
    161
    +0.11
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.58
    0.19
    111
    +0.55
    0.21
    419
    +0.11
    0.14
    Negative Logits
    ĸ
    -5.06
    Ī
    -5.06
    ¶
    -5.02
    ´
    -5.01
                ↵
    -4.97
    <|outofrange|>
    -4.97
    -4.97
    ↵
    -4.97
    -4.97
    -4.97
    POSITIVE LOGITS
    endent
    1.21
     seu
    1.20
    thems
    1.16
     cele
    1.16
     merit
    1.14
    emente
    1.12
    hang
    1.11
    hou
    1.08
     opinions
    1.08
    ainties
    1.07
    Activations Density 0.268%

    No Known Activations