© 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. 28548
    Prev
    Next
    INDEX
    Explanations

    programming or coding terminology and syntax

    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
    293
    +0.20
    1.1%
    198
    +0.13
    0.8%
    332
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    293
    +0.20
    0.12
    332
    +0.13
    0.05
    479
    +0.13
    0.02
    Negative Logits
    TypeDef
    -1.59
    "}](#
    -1.55
    icense
    -1.53
    inois
    -1.49
    uximab
    -1.42
    roup
    -1.40
    jcmm
    -1.38
    arton
    -1.30
     Photograph
    -1.30
     subpoena
    -1.26
    POSITIVE LOGITS
     respectively
    1.43
    ishly
    1.42
     accordingly
    1.37
    ursed
    1.34
    urally
    1.34
    ially
    1.33
     marks
    1.30
    !).
    1.26
    itively
    1.25
    andy
    1.25
    Activations Density 5.479%

    No Known Activations