© 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. 31145
    Prev
    Next
    INDEX
    Explanations

    descriptions of individuals and their actions

    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
    447
    +0.36
    1.8%
    393
    +0.22
    1.1%
    445
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.36
    0.29
    0
    +0.22
    0.16
    128
    +0.08
    0.15
    Negative Logits
    adelphia
    -0.68
    odcast
    -0.68
    wire
    -0.66
    ETS
    -0.63
    Wire
    -0.62
    pees
    -0.62
     Powered
    -0.62
    wired
    -0.62
    zens
    -0.60
    lucent
    -0.58
    POSITIVE LOGITS
     therefore
    1.05
     consequently
    1.00
     thus
    0.98
     hence
    0.96
     thereby
    0.95
     consequ
    0.81
     possibly
    0.75
     nuance
    0.74
     righteousness
    0.71
     thence
    0.71
    Activations Density 11.777%

    No Known Activations