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. 29538
    Prev
    Next
    INDEX
    Explanations

    phrases and expressions related to expectations and forecasts

    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
    444
    +0.12
    0.7%
    117
    +0.12
    0.7%
    500
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    117
    +0.12
    0.04
    148
    +0.12
    0.01
    444
    +0.11
    0.01
    Negative Logits
    §
    -2.96
    Ļª
    -2.79
    Ļ
    -2.72
    IJ
    -2.65
    Īĺ
    -2.61
    ¡
    -2.59
    «
    -2.57
    ĥ½
    -2.53
    ¤
    -2.52
    Ń
    -2.50
    POSITIVE LOGITS
     result
    1.65
    latest
    1.54
     soon
    1.46
     fresh
    1.45
     white
    1.44
     goodbye
    1.43
     fare
    1.42
     dark
    1.41
     passage
    1.40
     however
    1.40
    Activations Density 0.254%

    No Known Activations