Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 1-RES-SM
    6. 26895
    Prev
    Next
    INDEX
    Explanations

    references to dermatology and skin-related medical terms

    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
    111
    +0.14
    0.8%
    391
    +0.13
    0.7%
    450
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    391
    +0.14
    0.01
    43
    +0.13
    0.02
    304
    +0.11
    0.02
    Negative Logits
     ya
    -1.68
     temper
    -1.55
    pecially
    -1.51
    verting
    -1.48
     turn
    -1.43
     men
    -1.41
     sworn
    -1.40
     sudden
    -1.39
     **
    -1.39
     wipe
    -1.38
    POSITIVE LOGITS
    ière
    1.92
    icile
    1.87
    sheet
    1.76
    esse
    1.74
    ridges
    1.71
    ionization
    1.70
    ű
    1.64
    esan
    1.59
    es
    1.57
    auer
    1.53
    Activations Density 0.102%

    No Known Activations