© 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. 3616
    Prev
    Next
    INDEX
    Explanations

    links to various online resources and computer programming instructions

    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
    640
    +0.32
    1.1%
    393
    +0.20
    0.7%
    163
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.32
    0.07
    257
    +0.20
    0.06
    284
    +0.15
    0.06
    Negative Logits
    Ļ½
    -1.15
    tsky
    -1.14
    anooga
    -1.11
    ¬¼
    -1.08
    ĪĴ
    -1.05
     âī¡
    -1.01
    IER
    -1.01
     behavi
    -0.99
    ortunately
    -0.98
    ¯¯
    -0.94
    POSITIVE LOGITS
    youtube
    1.56
    google
    1.54
    example
    1.48
    ItemTracker
    1.48
    dat
    1.46
    mk
    1.46
    debian
    1.45
    planet
    1.42
    amazon
    1.38
    github
    1.37
    Activations Density 0.429%

    No Known Activations