© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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 Local
    9. 6-RES_SCL-AJT
    11. 33290
    Prev
    Next
    INDEX
    Explanations

    mentions of personal experiences and emotions

    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
    •Layer 6 UMAP region: Mostly-local line at bottom - local

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.23
    1.8%
    64
    +0.19
    1.5%
    87
    +0.18
    1.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.23
    -0.04
    756
    +0.19
    -0.07
    283
    +0.18
    0.02
    Negative Logits
     sucks
    -0.39
     Appears
    -0.36
    pires
    -0.36
    requires
    -0.36
    )?
    -0.35
     WATCH
    -0.34
     Goo
    -0.34
     depends
    -0.34
     =================================
    -0.34
     comes
    -0.34
    POSITIVE LOGITS
     enthusi
    0.39
     tended
    0.37
     earlier
    0.35
     cumbers
    0.35
     arranging
    0.35
     initially
    0.34
    Was
    0.34
     hoped
    0.33
     beforehand
    0.32
    Had
    0.32
    Activations Density 4.708%

    No Known Activations