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. 29100
    Prev
    Next
    INDEX
    Explanations

    references to various sports

    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
    350
    +0.14
    0.8%
    389
    +0.14
    0.8%
    155
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    389
    +0.14
    0.03
    155
    +0.14
    0.03
    112
    +0.13
    0.02
    Negative Logits
    rowser
    -1.73
     manner
    -1.61
     miscar
    -1.58
     fois
    -1.52
    nlm
    -1.50
     wake
    -1.50
     bye
    -1.48
     fore
    -1.47
     meu
    -1.47
     libc
    -1.44
    POSITIVE LOGITS
    smen
    2.97
    icity
    2.31
    ivity
    2.21
    sw
    2.20
    sm
    2.16
    iest
    2.13
    ivation
    2.12
    ive
    2.08
    iveness
    2.01
    ivated
    1.96
    Activations Density 0.207%

    No Known Activations