© 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. 13405
    Prev
    Next
    INDEX
    Explanations

    formatting and structure in text-based content, possibly including code or programming language elements

    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
    +0.18
    1.1%
    337
    +0.17
    1.0%
    25
    +0.16
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.18
    0.06
    444
    +0.17
    0.10
    337
    +0.16
    0.08
    Negative Logits
    leftarrow
    -1.31
    Affirmed
    -1.20
    use
    -1.19
    rVert
    -1.15
     Serve
    -1.14
     deletions
    -1.12
    namespace
    -1.12
    age
    -1.10
     namespace
    -1.09
    lookup
    -1.08
    POSITIVE LOGITS
    nes
    1.51
    terness
    1.47
     $|\
    1.32
     whilst
    1.30
    hene
    1.27
     waking
    1.26
    mouth
    1.26
     blindness
    1.26
    fections
    1.26
     consciousness
    1.25
    Activations Density 2.892%

    No Known Activations