© 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. 2-RES-SM
    11. 9791
    Prev
    Next
    INDEX
    Explanations

    repetitive terms indicating the current state or frequency of events

    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
    423
    +0.13
    0.7%
    274
    +0.11
    0.6%
    48
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.13
    0.06
    335
    +0.11
    0.05
    33
    +0.10
    0.05
    Negative Logits
    ?”
    -1.81
    ibles
    -1.76
    \]\].
    -1.74
     preach
    -1.66
    .\
    -1.60
    ”?
    -1.59
     wonders
    -1.58
    "?
    -1.54
    .";
    -1.53
    ?).
    -1.52
    POSITIVE LOGITS
     however
    1.78
    idden
    1.61
    however
    1.51
    enso
    1.49
     Generated
    1.42
    ourse
    1.40
    ert
    1.32
    ische
    1.31
     commentators
    1.31
     these
    1.30
    Activations Density 0.422%

    No Known Activations