© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 29694
    Prev
    Next
    INDEX
    Explanations

    negative references to user experiences or data related to products and services

    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
    235
    +0.16
    0.9%
    326
    +0.14
    0.8%
    98
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.16
    0.06
    220
    +0.14
    0.04
    259
    +0.13
    0.03
    Negative Logits
    Ļ
    -2.01
    ĩ
    -2.00
    ¤
    -1.98
    ·
    -1.95
    ĸ
    -1.95
    ¸
    -1.95
    ĻĤ
    -1.93
    Ĥ
    -1.89
    ½
    -1.86
    ¿½
    -1.86
    POSITIVE LOGITS
    lic
    1.50
    igo
    1.49
    icides
    1.48
    ALS
    1.45
     counties
    1.43
     immunity
    1.43
     Province
    1.42
    ename
    1.41
     County
    1.39
     poisoning
    1.39
    Activations Density 0.661%

    No Known Activations