© 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. 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. 40387
    Prev
    Next
    INDEX
    Explanations

    phrases related to ethics and society

    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
    +0.21
    0.9%
    7
    +0.14
    0.6%
    692
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    692
    +0.21
    0.05
    51
    +0.14
    0.13
    370
    +0.13
    0.06
    Negative Logits
    Ur
    -0.74
     Nicarag
    -0.73
    ãĥīãĥ©
    -0.73
     unspecified
    -0.70
    Iraq
    -0.70
    SPONSORED
    -0.69
    çīĪ
    -0.69
     Urug
    -0.68
    Offic
    -0.67
     Nord
    -0.67
    POSITIVE LOGITS
     nowadays
    1.09
     inherently
    1.05
    ateur
    1.01
     notoriously
    0.85
    ometimes
    0.84
     typically
    0.83
    itect
    0.82
     usually
    0.81
     ALWAYS
    0.81
    ynthesis
    0.81
    Activations Density 23.112%

    No Known Activations