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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29298
    Prev
    Next
    INDEX
    Explanations

    links related to sharing content or objects within digital communications

    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
    445
    +0.14
    0.8%
    329
    +0.13
    0.7%
    460
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    329
    +0.14
    0.04
    445
    +0.13
    0.03
    82
    +0.12
    0.04
    Negative Logits
    ising
    -1.56
    osing
    -1.51
    olis
    -1.51
     depended
    -1.49
    atching
    -1.44
    oden
    -1.44
    otyped
    -1.41
    abi
    -1.39
    oping
    -1.34
     weren
    -1.32
    POSITIVE LOGITS
     NOTICE
    1.67
    ¬
    1.61
    Īĺ
    1.55
    «
    1.50
     COPYRIGHT
    1.49
     "</
    1.47
    ·
    1.46
     YOU
    1.44
     caption
    1.41
     poster
    1.41
    Activations Density 0.737%

    No Known Activations