© 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. 12909
    Prev
    Next
    INDEX
    Explanations

    phrases indicating an increase or rise

    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
    +0.15
    0.9%
    502
    +0.14
    0.8%
    348
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    41
    +0.15
    0.05
    164
    +0.14
    0.05
    502
    +0.12
    0.05
    Negative Logits
    ized
    -1.77
    ization
    -1.67
    ize
    -1.67
    ties
    -1.53
    ators
    -1.51
    izer
    -1.47
    izers
    -1.47
    plain
    -1.41
    OID
    -1.39
     intent
    -1.39
    POSITIVE LOGITS
    hol
    2.49
    scaling
    2.45
    bringing
    2.34
    holding
    2.33
    holds
    2.31
    hill
    2.25
    graded
    2.13
    raised
    2.11
    vote
    2.10
    uncture
    2.05
    Activations Density 0.186%

    No Known Activations