© 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. 28760
    Prev
    Next
    INDEX
    Explanations

    the occurrence of thoughts or contemplative states

    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
    13
    +0.12
    0.7%
    115
    +0.12
    0.7%
    348
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.12
    0.03
    471
    +0.12
    0.02
    115
    +0.11
    0.02
    Negative Logits
    ĭ
    -2.22
    ¨
    -2.20
    ľĵ
    -2.18
    ĥ
    -2.14
    ľ
    -1.94
    ·¸
    -1.94
    ·
    -1.92
    Ģ
    -1.92
    ĻĤ
    -1.90
    ķ
    -1.83
    POSITIVE LOGITS
     aloud
    1.98
    about
    1.88
    fully
    1.84
    lessly
    1.84
     about
    1.70
    fulness
    1.66
     goodbye
    1.63
     About
    1.60
    ful
    1.56
     prov
    1.49
    Activations Density 0.163%

    No Known Activations