© 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 EvalsExportsinterp-engineNEWAPI 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. 32258
    Prev
    Next
    INDEX
    Explanations

    summaries or conclusions

    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
    39
    +0.10
    0.5%
    274
    +0.10
    0.5%
    447
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.10
    0.08
    393
    +0.10
    0.19
    698
    +0.09
    0.06
    Negative Logits
    iae
    -0.66
    ali
    -0.60
    iaries
    -0.58
    iaz
    -0.57
    roid
    -0.55
    keys
    -0.54
     photographed
    -0.54
    jab
    -0.53
    aline
    -0.53
     proxies
    -0.53
    POSITIVE LOGITS
     succinct
    0.94
    Ĥª
    0.76
     summed
    0.75
     concise
    0.71
    ĺħ
    0.67
     summarizes
    0.65
     neatly
    0.63
     nicely
    0.61
    tten
    0.61
     usefulness
    0.60
    Activations Density 7.080%

    No Known Activations