© 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. 2-RES-SM
    11. 31302
    Prev
    Next
    INDEX
    Explanations

    words beginning with "im-", particularly those related to immediacy or significance like "immediately", "imminent", "improved", "impossible", etc.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @aml7990

    variations of the verb "imagine."

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    405
    +0.13
    0.7%
    81
    +0.12
    0.6%
    230
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.02
    391
    +0.12
    0.03
    177
    +0.11
    0.02
    Negative Logits
    shore
    -1.74
    oming
    -1.63
    rait
    -1.56
    MSO
    -1.51
     latter
    -1.49
     likelihood
    -1.48
    icated
    -1.46
    ENTIAL
    -1.46
    mouth
    -1.42
    irector
    -1.42
    POSITIVE LOGITS
    perial
    2.24
    ulator
    1.82
    agine
    1.73
    ulators
    1.72
    ulation
    1.69
    ulative
    1.68
    ulus
    1.67
    mediately
    1.62
    plify
    1.61
    mediate
    1.59
    Activations Density 0.140%

    No Known Activations