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. 29443
    Prev
    Next
    INDEX
    Explanations

    types and categories of objects or substances

    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
    17
    +0.24
    1.6%
    47
    +0.21
    1.4%
    198
    +0.19
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.24
    0.02
    47
    +0.21
    0.14
    198
    +0.19
    0.09
    Negative Logits
    ª
    -2.86
    ¦
    -2.85
    ©
    -2.74
    ¸
    -2.73
    §
    -2.68
    Ĩ
    -2.64
    ¨
    -2.62
    Ļ
    -2.62
    ¿½
    -2.61
    ¯
    -2.58
    POSITIVE LOGITS
    ents
    1.42
    ouble
    1.39
    arity
    1.38
    uple
    1.38
     footprint
    1.38
     Rapids
    1.37
    hedron
    1.36
    orphism
    1.34
    ellation
    1.34
     context
    1.34
    Activations Density 2.183%

    No Known Activations