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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 13933
    Prev
    Next
    INDEX
    Explanations

    references to geek culture and related topics, including science fiction, comic books, and technology

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    48
    +0.15
    0.6%
    385
    +0.13
    0.5%
    578
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.17
    326
    +0.13
    0.13
    568
    +0.12
    0.07
    Negative Logits
    undai
    -0.82
     reconc
    -0.74
     separating
    -0.73
     Chao
    -0.70
     OB
    -0.70
    estinal
    -0.67
     chlorine
    -0.66
    inates
    -0.65
    eele
    -0.65
    ouf
    -0.65
    POSITIVE LOGITS
    core
    1.02
    Dad
    1.02
    arthed
    0.99
    roots
    0.89
     community
    0.87
    adelphia
    0.86
     fandom
    0.83
    istani
    0.82
    Tang
    0.82
    Leaks
    0.81
    Activations Density 2.597%

    No Known Activations