© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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 Local
    9. 6-RES_SCL-AJT
    11. 31366
    Prev
    Next
    INDEX
    Explanations

    names and partial names starting with "Rom" or "Ron."

    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
    Neuron Alignment
    Index
    Value
    % of L₁
    178
    +0.12
    0.6%
    588
    +0.08
    0.4%
    681
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.22
    11
    +0.08
    0.07
    659
    +0.08
    0.11
    Negative Logits
    gettable
    -0.66
    insula
    -0.59
     tongues
    -0.58
     attendant
    -0.57
    velength
    -0.56
    Ļ½
    -0.56
     sled
    -0.55
     exha
    -0.54
     graded
    -0.54
     Rouhani
    -0.53
    POSITIVE LOGITS
    ulus
    0.93
    ijn
    0.86
    antics
    0.83
    anes
    0.81
    atic
    0.80
    ace
    0.79
    olution
    0.77
    ances
    0.74
    endor
    0.74
    agna
    0.73
    Activations Density 7.243%

    No Known Activations