© 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
    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. 40581
    Prev
    Next
    INDEX
    Explanations

    references to quotes and sources, particularly with numbers like ISBNs

    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
    308
    +0.12
    0.5%
    306
    +0.10
    0.4%
    61
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    444
    +0.12
    0.04
    326
    +0.10
    0.13
    212
    +0.10
    0.04
    Negative Logits
    ĪĴ
    -0.96
    abal
    -0.88
    uliffe
    -0.86
    pez
    -0.80
    ulhu
    -0.80
    phabet
    -0.78
    obyl
    -0.76
    ambers
    -0.74
    rera
    -0.74
    uty
    -0.72
    POSITIVE LOGITS
    76561
    0.98
     978
    0.82
     ISBN
    0.77
    ãĥĻ
    0.76
    ://
    0.76
    978
    0.76
    ãĥĭ
    0.74
    ãĥł
    0.73
    ãĥ
    0.73
    UTF
    0.71
    Activations Density 3.215%

    No Known Activations