© 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. 2-RES-SM
    11. 30684
    Prev
    Next
    INDEX
    Explanations

    the word "which" in various contexts

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    51
    +0.13
    0.7%
    310
    +0.12
    0.7%
    169
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    51
    +0.13
    0.03
    149
    +0.12
    0.02
    306
    +0.11
    0.02
    Negative Logits
    us
    -1.53
     center
    -1.48
    .",
    -1.42
    '):
    -1.41
     centre
    -1.39
    .\[[@
    -1.33
     ...,
    -1.29
    ",
    -1.27
     applications
    -1.27
     \[*
    -1.26
    POSITIVE LOGITS
     same
    2.01
    same
    1.70
     SAME
    1.54
    ighs
    1.51
    rael
    1.51
    legal
    1.49
    GPL
    1.46
    soever
    1.46
    falls
    1.45
    friends
    1.40
    Activations Density 0.076%

    No Known Activations