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. 2-RES-SM
    6. 11798
    Prev
    Next
    INDEX
    Explanations

    terms related to commercial enterprises and activities

    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
    156
    +0.20
    1.2%
    186
    +0.13
    0.7%
    208
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    208
    +0.20
    0.03
    418
    +0.13
    0.00
    421
    +0.12
    0.01
    Negative Logits
     slept
    -1.63
    kan
    -1.62
     honored
    -1.52
     "!"
    -1.48
    onto
    -1.43
     unconscious
    -1.38
    chitz
    -1.37
    uo
    -1.37
    ĨĴ
    -1.36
     sleep
    -1.35
    POSITIVE LOGITS
    isation
    2.23
    ising
    2.18
    ization
    2.12
    ised
    1.89
    ized
    1.88
    isations
    1.87
    izations
    1.83
     establishments
    1.82
    izing
    1.82
    isers
    1.73
    Activations Density 0.123%

    No Known Activations