Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 13356
    Prev
    Next
    INDEX
    Explanations

    phrases related to conspiracy theories or controversial declarations, potentially involving different words like "plot," "plan," "strategy," or "declaration."

    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
    55
    +0.11
    0.5%
    488
    +0.11
    0.5%
    6
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.11
    -0.00
    107
    +0.11
    0.12
    48
    +0.10
    0.16
    Negative Logits
    undai
    -0.81
    eatures
    -0.80
     reluct
    -0.80
     scrut
    -0.78
    EStreamFrame
    -0.77
     satell
    -0.76
     redes
    -0.76
    raviolet
    -0.73
    owship
    -0.73
    orsche
    -0.73
    POSITIVE LOGITS
    âĢİ
    0.92
    çļĦ
    0.89
    âĢ
    0.84
     istg
    0.81
    ÙĨ
    0.80
    ãĥ¼
    0.80
    н
    0.78
    ÏĤ
    0.78
    (-
    0.75
    в
    0.74
    Activations Density 14.745%

    No Known Activations