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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 11649
    Prev
    Next
    INDEX
    Explanations

    occurrences of sporting events and match details

    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
    478
    +0.18
    1.0%
    69
    +0.13
    0.7%
    442
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.18
    0.17
    238
    +0.13
    0.10
    362
    +0.11
    0.12
    Negative Logits
    liness
    -1.63
     sense
    -1.55
    veolar
    -1.45
    otype
    -1.43
    ção
    -1.42
     '%
    -1.41
    theless
    -1.40
     invariant
    -1.40
     picture
    -1.40
     inactivated
    -1.39
    POSITIVE LOGITS
    ¦
    3.43
    ĥ½
    3.34
    IJ
    3.30
    ĻĤ
    3.27
    ľĵ
    3.20
    ¼
    3.10
    °
    3.06
    ¶
    3.06
    ³
    3.03
    Ļª
    3.03
    Activations Density 1.869%

    No Known Activations