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. 29572
    Prev
    Next
    INDEX
    Explanations

    references to fighting or struggle in various contexts

    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
    156
    +0.16
    0.9%
    481
    +0.14
    0.8%
    304
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.16
    0.04
    56
    +0.14
    0.00
    304
    +0.13
    0.04
    Negative Logits
     vez
    -1.74
    blogger
    -1.69
    TRODUCTION
    -1.57
     habl
    -1.55
     valuable
    -1.51
    ureus
    -1.50
     handy
    -1.48
     aid
    -1.44
     privileged
    -1.41
     purchased
    -1.38
    POSITIVE LOGITS
    ground
    2.22
    hammer
    2.13
    grounds
    2.11
    oon
    2.03
     against
    2.01
    lord
    1.98
    oons
    1.78
    lords
    1.77
    front
    1.76
     scenes
    1.74
    Activations Density 0.383%

    No Known Activations