© 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. 4686
    Prev
    Next
    INDEX
    Explanations

    chemical compounds related to fermentation and food processing

    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
    582
    +0.14
    0.6%
    152
    +0.11
    0.5%
    23
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.14
    0.23
    659
    +0.11
    0.10
    683
    +0.10
    0.09
    Negative Logits
    merce
    -0.79
    ¥ŀ
    -0.74
    ISO
    -0.74
    uracy
    -0.73
     accrued
    -0.72
    rising
    -0.72
    PDATE
    -0.71
    edience
    -0.70
    ipolar
    -0.69
     inev
    -0.69
    POSITIVE LOGITS
    ength
    1.14
    witz
    1.11
    enger
    1.08
    mann
    1.02
    aer
    0.99
    mas
    0.98
    ibrary
    0.96
    arin
    0.96
    beck
    0.93
    atan
    0.93
    Activations Density 8.144%

    No Known Activations