Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11999
    Prev
    Next
    INDEX
    Explanations

    elements of structured data or code 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
    289
    +0.12
    0.7%
    481
    +0.10
    0.6%
    428
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.12
    0.08
    67
    +0.10
    0.07
    351
    +0.10
    0.07
    Negative Logits
     fu
    -1.67
    )$\
    -1.66
    ises
    -1.56
    )',
    -1.49
     analys
    -1.39
    ()),
    -1.39
     emergencies
    -1.37
    ère
    -1.34
    adays
    -1.34
    ón
    -1.32
    POSITIVE LOGITS
    ·
    4.33
    ĭ
    4.21
    ©
    4.11
    ¦
    4.07
    ĻĤ
    4.04
    Ĥ¬
    4.04
    Ģ
    4.03
    ±
    4.03
    į
    3.97
    ŀ
    3.96
    Activations Density 0.324%

    No Known Activations