© 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. 12375
    Prev
    Next
    INDEX
    Explanations

    expressions of strong emotions or sentiments

    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
    78
    +0.13
    0.7%
    218
    +0.11
    0.6%
    21
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.13
    0.09
    377
    +0.11
    0.09
    202
    +0.11
    0.02
    Negative Logits
    ations
    -1.79
    ipe
    -1.55
    gas
    -1.54
    etype
    -1.52
    continue
    -1.47
    any
    -1.45
    Ħ
    -1.45
     happening
    -1.43
    elihood
    -1.43
    PLC
    -1.43
    POSITIVE LOGITS
     aware
    2.91
     happy
    2.70
     anxious
    2.68
     acquainted
    2.66
     glad
    2.58
     pleased
    2.48
     thankful
    2.44
     conscious
    2.41
     grateful
    2.41
     confident
    2.38
    Activations Density 0.932%

    No Known Activations