© 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. 4-RES-SM
    11. 14758
    Prev
    Next
    INDEX
    Explanations

    modal verbs indicating possibility or uncertainty

    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
    290
    +0.13
    0.7%
    391
    +0.12
    0.7%
    461
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    290
    +0.13
    0.06
    96
    +0.12
    0.04
    391
    +0.11
    0.07
    Negative Logits
    ļ
    -1.86
    ĻĤ
    -1.65
    Ĭ
    -1.60
    ¢
    -1.55
    Ģ
    -1.55
    olution
    -1.55
    etus
    -1.54
    Ĥ¬
    -1.51
    º
    -1.50
    ires
    -1.49
    POSITIVE LOGITS
    haps
    2.41
    ily
    2.12
    hem
    2.10
     well
    2.08
     someday
    1.98
     be
    1.98
    oral
    1.97
    ier
    1.90
    onna
    1.89
    pole
    1.83
    Activations Density 0.246%

    No Known Activations