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. 28821
    Prev
    Next
    INDEX
    Explanations

    references to women's health and wellbeing 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
    271
    +0.20
    1.2%
    261
    +0.16
    1.0%
    186
    +0.16
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.20
    0.08
    441
    +0.16
    -0.02
    271
    +0.16
    -0.06
    Negative Logits
    sheet
    -1.54
     edition
    -1.49
    statement
    -1.41
     offers
    -1.41
     recogn
    -1.40
     defenses
    -1.38
     itself
    -1.38
     won
    -1.36
     findViewById
    -1.35
     competence
    -1.30
    POSITIVE LOGITS
    ĻĤ
    2.91
    Īĺ
    2.54
    ¿½
    2.41
    IJ
    2.36
    »¿
    2.24
    ĭ
    2.21
    à¯į
    2.17
    º
    2.17
    Ļª
    2.12
    ħ
    2.05
    Activations Density 6.194%

    No Known Activations