© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12051
    Prev
    Next
    INDEX
    Explanations

    references to specific characters and events within a narrative context, particularly involving heroes and villains

    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
    21
    +0.40
    2.7%
    369
    +0.19
    1.3%
    410
    +0.13
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.40
    0.11
    369
    +0.19
    0.10
    98
    +0.13
    0.07
    Negative Logits
    »¿
    -1.59
    ĥ½
    -1.40
    vironment
    -1.40
    fections
    -1.39
     apparatus
    -1.39
    ections
    -1.38
     manner
    -1.36
     process
    -1.35
     tails
    -1.34
    xture
    -1.34
    POSITIVE LOGITS
    ’
    1.73
    &
    1.45
    annie
    1.42
     meant
    1.42
    ulle
    1.35
     fans
    1.33
    ollo
    1.32
     Motors
    1.31
    ‘
    1.31
    olla
    1.29
    Activations Density 1.213%

    No Known Activations