© 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. 12361
    Prev
    Next
    INDEX
    Explanations

    instances of the word "but."

    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
    316
    +0.13
    0.7%
    21
    +0.11
    0.6%
    96
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.13
    0.02
    316
    +0.11
    0.03
    146
    +0.11
    0.04
    Negative Logits
    ¸
    -3.44
    µ
    -3.21
    ·
    -2.97
    ¶
    -2.96
    ²
    -2.81
    Īĺ
    -2.80
    ³
    -2.80
    £
    -2.79
    ¹
    -2.75
    ½
    -2.72
    POSITIVE LOGITS
    tons
    3.04
    chers
    2.87
    ted
    2.70
    ler
    2.69
    cher
    2.60
    tered
    2.59
    ts
    2.51
    thead
    2.48
    tes
    2.46
    term
    2.45
    Activations Density 0.211%

    No Known Activations