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

    Neuronpedia

    APICircuit 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. 11599
    Prev
    Next
    INDEX
    Explanations

    instances of quotes or dialogue

    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
    +0.16
    0.9%
    6
    +0.15
    0.8%
    261
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.16
    0.04
    66
    +0.15
    0.04
    466
    +0.14
    0.04
    Negative Logits
     existed
    -1.57
    etition
    -1.56
     matters
    -1.52
    ests
    -1.51
     exists
    -1.50
    ository
    -1.47
     mattered
    -1.45
    fecture
    -1.42
    icial
    -1.41
     exist
    -1.40
    POSITIVE LOGITS
    ĭ
    2.64
    ®
    2.63
    ĻĤ
    2.40
    ¿½
    2.38
    ľĵ
    2.27
    °
    2.24
    ·
    2.19
    Ľ
    2.15
    ³
    2.13
    ¦
    2.10
    Activations Density 0.220%

    No Known Activations