© 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. 9002
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "the"

    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
    21
    +0.25
    1.5%
    450
    +0.12
    0.7%
    64
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    278
    +0.25
    0.02
    179
    +0.12
    0.02
    162
    +0.12
    0.01
    Negative Logits
    keit
    -1.88
    hin
    -1.71
    ment
    -1.56
    osse
    -1.56
    na
    -1.53
    ally
    -1.47
    ically
    -1.46
    ely
    -1.45
    itis
    -1.40
    ials
    -1.38
    POSITIVE LOGITS
     Apache
    1.84
     WARRANTY
    1.64
     Redistributions
    1.63
     WARRANTIES
    1.60
     CONTRIBUTORS
    1.58
     GNU
    1.53
     Source
    1.48
    7554
    1.48
     {@
    1.46
     GPL
    1.44
    Activations Density 0.036%

    No Known Activations