© 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. 27468
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "observed" and its variations related to observations or sightings

    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
    148
    +0.15
    0.8%
    314
    +0.13
    0.7%
    376
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.15
    0.01
    343
    +0.13
    0.03
    115
    +0.12
    0.02
    Negative Logits
    ĵ
    -4.22
    ĭ
    -4.21
    Īĺ
    -4.05
    ŀ
    -3.97
    «
    -3.94
    ½
    -3.92
    ģ
    -3.80
    Ľ
    -3.67
    ĸ
    -3.67
    Ģ
    -3.64
    POSITIVE LOGITS
     faint
    1.60
     finding
    1.59
    itch
    1.58
    initis
    1.57
     peaks
    1.53
    Errno
    1.52
    above
    1.52
     sight
    1.50
     phenomenon
    1.49
     seeing
    1.45
    Activations Density 0.139%

    No Known Activations