© 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. 45984
    Prev
    Next
    INDEX
    Explanations

    mentions of anonymous sources, information, and 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
    +0.21
    1.1%
    288
    +0.14
    0.8%
    385
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.21
    0.15
    201
    +0.14
    0.10
    669
    +0.11
    0.08
    Negative Logits
    NEY
    -0.76
    rehens
    -0.74
    neys
    -0.73
    enegger
    -0.72
    Ħ¢
    -0.69
    bourg
    -0.69
    union
    -0.68
    united
    -0.65
    master
    -0.63
    hews
    -0.63
    POSITIVE LOGITS
     informant
    0.81
     anonymous
    0.80
    onym
    0.78
     anonymity
    0.73
     anonymously
    0.72
     sourcing
    0.72
    ously
    0.72
     whistleblower
    0.69
     tip
    0.68
     narrator
    0.65
    Activations Density 8.551%

    No Known Activations