© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5997
    Prev
    Next
    INDEX
    Explanations

    phrases related to history and timelines

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases that indicate a duration or a specific point in time

    oai_token-act-pair · gpt-3.5-turboTriggered by @jordan
    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
    393
    +0.14
    0.5%
    558
    +0.14
    0.5%
    505
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    706
    +0.14
    0.08
    71
    +0.14
    0.03
    269
    +0.13
    0.04
    Negative Logits
    igible
    -0.87
    ucker
    -0.86
     exceptions
    -0.83
    inki
    -0.82
    soever
    -0.81
    ellow
    -0.81
    ouf
    -0.79
    ilities
    -0.79
    uties
    -0.78
    aug
    -0.77
    POSITIVE LOGITS
    since
    1.08
     countless
    1.03
     decades
    0.96
     YEAR
    0.94
     millennia
    0.92
     awhile
    0.90
    many
    0.89
     sometime
    0.88
     centuries
    0.86
     since
    0.86
    Activations Density 1.528%

    No Known Activations