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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 3766
    Prev
    Next
    INDEX
    Explanations

    mentions of medical and health-related terms and issues

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.30
    1.5%
    393
    +0.23
    1.1%
    202
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.30
    0.20
    546
    +0.23
    0.20
    393
    +0.10
    0.30
    Negative Logits
    ady
    -0.63
    ¬¼
    -0.61
    isi
    -0.58
    izens
    -0.58
    eny
    -0.57
    emaker
    -0.57
    orem
    -0.55
    é¾
    -0.54
    afe
    -0.54
    enses
    -0.54
    POSITIVE LOGITS
     yeah
    1.01
     namely
    0.90
     huh
    0.90
     etc
    0.88
     whereas
    0.82
     although
    0.79
     maybe
    0.77
     blah
    0.76
     frankly
    0.75
     mathemat
    0.74
    Activations Density 11.514%

    No Known Activations