© 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. 11318
    Prev
    Next
    INDEX
    Explanations

    terms related to negative experiences or consequences

    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
    477
    +0.13
    0.8%
    59
    +0.12
    0.7%
    501
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.02
    59
    +0.12
    0.03
    74
    +0.11
    0.03
    Negative Logits
     saf
    -1.52
     keyboard
    -1.48
     Weekly
    -1.47
     safety
    -1.46
     include
    -1.43
     shortest
    -1.36
     favorable
    -1.35
     other
    -1.35
     Week
    -1.34
     spare
    -1.31
    POSITIVE LOGITS
    ³
    3.17
    ±
    3.02
    ĻĤ
    2.95
    ŀ
    2.88
    ´
    2.73
    ·¸
    2.72
    ¡
    2.72
    İ
    2.69
    Ĥ
    2.68
    ģ
    2.65
    Activations Density 0.301%

    No Known Activations