Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 13605
    Prev
    Next
    INDEX
    Explanations

    detailed numerical data, including measurements, percentages, and specific values

    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
    +0.23
    1.0%
    311
    +0.19
    0.8%
    393
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.23
    0.19
    283
    +0.19
    0.03
    589
    +0.12
    0.07
    Negative Logits
    lehem
    -0.83
    erity
    -0.83
    blance
    -0.76
    cerning
    -0.73
    女
    -0.70
    Ü
    -0.68
    DragonMagazine
    -0.68
    soever
    -0.68
    Daily
    -0.66
    ItemImage
    -0.65
    POSITIVE LOGITS
    xff
    1.24
    xc
    1.01
    xe
    1.00
    xb
    0.99
    xa
    0.96
    xd
    0.96
    x
    0.94
    xes
    0.82
    644
    0.80
    uden
    0.78
    Activations Density 7.908%

    No Known Activations