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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 546
    Prev
    Next
    INDEX
    Explanations

    phrases related to historical events or facts

    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
    267
    +0.16
    0.6%
    339
    +0.14
    0.5%
    99
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.16
    0.05
    267
    +0.14
    0.05
    99
    +0.14
    0.05
    Negative Logits
    rom
    -1.08
    laws
    -0.94
    ilation
    -0.94
    vich
    -0.92
    bara
    -0.90
    ific
    -0.89
    esses
    -0.89
     Vish
    -0.89
    rament
    -0.87
    holes
    -0.87
    POSITIVE LOGITS
     conceived
    1.32
     intended
    1.14
     planted
    1.06
     opposed
    1.06
     plotted
    1.06
     slated
    1.05
     designed
    1.03
     formulated
    1.02
     coined
    1.01
     appre
    1.01
    Activations Density 0.627%

    No Known Activations