© 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. 40403
    Prev
    Next
    INDEX
    Explanations

    references to Edward Snowden and NSA spying activities

    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
    +0.18
    0.9%
    288
    +0.14
    0.7%
    571
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.18
    0.11
    554
    +0.14
    -0.07
    465
    +0.11
    0.05
    Negative Logits
    sembly
    -0.78
    rontal
    -0.72
    hetical
    -0.68
    reek
    -0.66
    ptive
    -0.65
    opathic
    -0.64
    ogun
    -0.64
    ombo
    -0.63
    ardon
    -0.63
    ournals
    -0.62
    POSITIVE LOGITS
     Snowden
    0.78
    DragonMagazine
    0.78
    den
    0.65
    çīĪ
    0.64
    ç·
    0.60
     bin
    0.58
    æľ
    0.58
    Jr
    0.58
    obyl
    0.56
    loo
    0.56
    Activations Density 5.592%

    No Known Activations