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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 29029
    Prev
    Next
    INDEX
    Explanations

    phrases related to software licensing and usage rights

    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
    323
    +0.15
    0.8%
    299
    +0.12
    0.7%
    74
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    222
    +0.15
    0.03
    323
    +0.12
    0.02
    74
    +0.11
    0.03
    Negative Logits
    ĨĴ
    -1.90
    olith
    -1.56
    oden
    -1.52
    ®
    -1.51
    abi
    -1.48
    olta
    -1.48
    Overall
    -1.45
    epid
    -1.44
    ¯
    -1.44
    earing
    -1.42
    POSITIVE LOGITS
    ↵Č
    1.69
     rights
    1.69
     appellant
    1.68
     xe
    1.65
     license
    1.65
     lic
    1.65
     attorney
    1.53
     hereinafter
    1.52
     jointly
    1.50
     Âĵ
    1.50
    Activations Density 0.135%

    No Known Activations