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. 12054
    Prev
    Next
    INDEX
    Explanations

    phrases related to file encoding and XML structure

    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
    460
    +0.13
    0.7%
    74
    +0.12
    0.7%
    76
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    74
    +0.13
    0.02
    316
    +0.12
    0.02
    253
    +0.12
    0.02
    Negative Logits
    wards
    -1.62
    ...](
    -1.62
    ”.
    -1.50
    after
    -1.49
    â̦”
    -1.47
    vie
    -1.46
    .’”
    -1.45
    .”
    -1.45
    etically
    -1.44
    ĩ
    -1.43
    POSITIVE LOGITS
    bourg
    1.53
     thickness
    1.37
    ación
    1.33
     corner
    1.32
    heit
    1.32
     Symphony
    1.29
     note
    1.29
     capacity
    1.28
     translation
    1.25
    елÑĮ
    1.25
    Activations Density 0.120%

    No Known Activations