© 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. 28540
    Prev
    Next
    INDEX
    Explanations

    programming-related terms and structures

    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
    342
    +0.14
    0.8%
    302
    +0.13
    0.8%
    160
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.14
    0.14
    49
    +0.13
    0.14
    408
    +0.13
    0.08
    Negative Logits
    ¬
    -1.70
    eness
    -1.63
    '):
    -1.55
    ¾
    -1.51
     precautions
    -1.48
    Īĺ
    -1.47
    ?’
    -1.44
    '),
    -1.44
    ĨĴ
    -1.37
     vent
    -1.35
    POSITIVE LOGITS
    orectal
    1.63
    arcoma
    1.59
    atte
    1.58
    áĢº
    1.53
    orta
    1.49
    amsbsy
    1.49
    orter
    1.48
    lei
    1.45
    amssymb
    1.45
    phia
    1.45
    Activations Density 4.070%

    No Known Activations