© 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. 1-RES-SM
    11. 24183
    Prev
    Next
    INDEX
    Explanations

    words related to service providers or entities involved in various forms of assistance

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.21
    1.2%
    144
    +0.13
    0.7%
    73
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    73
    +0.21
    0.05
    144
    +0.13
    0.05
    450
    +0.11
    0.04
    Negative Logits
    ¿½
    -2.37
    ĥ½
    -2.09
    ı
    -1.66
    rious
    -1.55
    ĥ
    -1.53
    ĻĤ
    -1.51
    ĵ
    -1.49
     Welsh
    -1.48
     shortening
    -1.47
    ©
    -1.44
    POSITIVE LOGITS
    pective
    2.39
    ue
    2.23
    orbent
    2.13
    clerosis
    2.06
    pace
    2.04
    helf
    1.99
    chaft
    1.93
    port
    1.88
    olutions
    1.86
    enthal
    1.84
    Activations Density 0.219%

    No Known Activations