© 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. 29838
    Prev
    Next
    INDEX
    Explanations

    numerical references, including URLs and associated metadata

    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
    169
    +0.13
    0.7%
    443
    +0.10
    0.6%
    110
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.13
    0.05
    110
    +0.10
    0.04
    169
    +0.10
    0.06
    Negative Logits
     kidding
    -1.50
    ahu
    -1.46
    rogens
    -1.41
    ]{.
    -1.41
    rogen
    -1.41
    vere
    -1.37
    ?)
    -1.33
     soon
    -1.32
    ...]
    -1.28
     late
    -1.27
    POSITIVE LOGITS
    essages
    1.81
     instance
    1.50
    ession
    1.46
    untime
    1.43
    ETHOD
    1.41
    LETE
    1.39
    gage
    1.37
    iewicz
    1.36
    unction
    1.36
    orro
    1.36
    Activations Density 0.395%

    No Known Activations