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. 28961
    Prev
    Next
    INDEX
    Explanations

    textual elements of dialogue or conversation

    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
    161
    +0.18
    1.0%
    204
    +0.16
    0.9%
    233
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.18
    0.16
    161
    +0.16
    0.05
    342
    +0.15
    0.07
    Negative Logits
    encing
    -1.49
     "(
    -1.41
    ences
    -1.39
    dig
    -1.30
    ând
    -1.29
     lives
    -1.29
     Domain
    -1.27
     "%
    -1.26
     ky
    -1.25
    vere
    -1.24
    POSITIVE LOGITS
    ľĵ
    3.70
    ©
    3.62
    ħ
    3.61
    ı
    3.56
    IJ
    3.55
    ĺ
    3.53
    ¨
    3.50
    ŀ
    3.43
    Ł
    3.42
    ª
    3.40
    Activations Density 7.784%

    No Known Activations