© 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. 1-RES-SM
    11. 28007
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "purpose" 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    158
    +0.13
    0.7%
    205
    +0.13
    0.7%
    376
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    158
    +0.13
    0.02
    205
    +0.13
    0.02
    464
    +0.12
    0.02
    Negative Logits
    gren
    -1.65
    anie
    -1.55
    uploads
    -1.55
    ↵³³
    -1.50
    dimen
    -1.46
    meg
    -1.45
    driv
    -1.43
    ietz
    -1.41
    innen
    -1.40
    opin
    -1.39
    POSITIVE LOGITS
    ful
    2.10
    ī
    1.87
    ory
    1.63
    fully
    1.63
     oriented
    1.59
    İ
    1.57
    keepers
    1.52
    º
    1.51
    orted
    1.49
    inib
    1.47
    Activations Density 0.073%

    No Known Activations