© 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. 11382
    Prev
    Next
    INDEX
    Explanations

    references to emotions and subjective experiences

    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
    197
    +0.15
    0.8%
    161
    +0.14
    0.8%
    183
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.15
    0.06
    36
    +0.14
    0.08
    243
    +0.13
    0.07
    Negative Logits
    Ī
    -2.83
    ĻĤ
    -2.72
    Ĵ
    -2.69
    Ļª
    -2.67
    £
    -2.63
    ī
    -2.59
    Ł
    -2.57
    ľĵ
    -2.56
    ĩ
    -2.55
    ŀ
    -2.53
    POSITIVE LOGITS
    _)
    1.51
     happening
    1.49
    ?_
    1.40
    )];
    1.40
     happens
    1.39
     happened
    1.36
    )_
    1.36
     ||
    1.36
    :]
    1.35
    Errno
    1.31
    Activations Density 1.146%

    No Known Activations