© 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. 40888
    Prev
    Next
    INDEX
    Explanations

    dates and time-related information

    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
    447
    +0.16
    0.8%
    310
    +0.11
    0.6%
    173
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.16
    0.13
    223
    +0.11
    0.11
    310
    +0.11
    0.11
    Negative Logits
    speaking
    -0.69
    atem
    -0.68
    Interstitial
    -0.66
     constitu
    -0.65
    desc
    -0.64
    ioch
    -0.64
     gifted
    -0.62
    Reviewer
    -0.59
     VIDEOS
    -0.58
    dyl
    -0.57
    POSITIVE LOGITS
     2017
    0.93
     2015
    0.89
     2019
    0.85
     2016
    0.85
     2018
    0.84
     2021
    0.84
     Madness
    0.82
     2014
    0.81
    bug
    0.80
    fair
    0.79
    Activations Density 7.834%

    No Known Activations