© 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. 29905
    Prev
    Next
    INDEX
    Explanations

    instances of the word "fix" and its variations related to troubleshooting or resolving issues

    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
    354
    +0.15
    0.9%
    270
    +0.12
    0.7%
    249
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    62
    +0.15
    0.01
    354
    +0.12
    0.02
    490
    +0.12
    0.02
    Negative Logits
    Ī
    -3.01
    Īĺ
    -2.78
    ĥ
    -2.77
    ¢
    -2.76
    ©
    -2.69
    Ĥ
    -2.68
    §
    -2.67
    Ļ
    -2.52
    Ĥ¬
    -2.51
    £
    -2.47
    POSITIVE LOGITS
     flaws
    1.88
    ention
    1.87
     problems
    1.81
     defects
    1.80
     deficiencies
    1.70
     issues
    1.69
    gered
    1.67
     bugs
    1.66
    issues
    1.65
    tons
    1.64
    Activations Density 0.146%

    No Known Activations