© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29942
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "used" in various contexts

    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
    69
    +0.12
    0.7%
    215
    +0.12
    0.7%
    252
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    215
    +0.12
    0.05
    398
    +0.12
    0.04
    287
    +0.11
    0.04
    Negative Logits
    mit
    -1.77
    ujah
    -1.70
    edly
    -1.52
    »¿
    -1.45
     departure
    -1.40
     ')
    -1.38
    à°¿
    -1.38
    FFIR
    -1.36
     Presidente
    -1.36
    ypse
    -1.33
    POSITIVE LOGITS
     alone
    1.84
    ries
    1.75
     extensively
    1.72
     mostly
    1.67
    etting
    1.60
     sparing
    1.58
     primarily
    1.52
    oline
    1.52
    orm
    1.51
    ogg
    1.49
    Activations Density 0.141%

    No Known Activations