© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 28734
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate involvement or participation 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
    282
    +0.16
    0.9%
    186
    +0.14
    0.8%
    485
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    282
    +0.16
    0.08
    69
    +0.14
    0.08
    440
    +0.13
    0.05
    Negative Logits
    heads
    -1.91
    head
    -1.76
    ogg
    -1.66
    stown
    -1.66
    weights
    -1.61
    setminus
    -1.56
     again
    -1.51
    ctin
    -1.49
    isters
    -1.49
    matter
    -1.45
    POSITIVE LOGITS
    »¿
    2.64
    ĥ½
    2.31
    ¿
    2.26
    ↵
    2.23
    ↵
    2.23
    ↵↵
    2.23
              ↵
    2.23
    2.23
     ↵
    2.23
    2.23
    Activations Density 0.542%

    No Known Activations