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. 45417
    Prev
    Next
    INDEX
    Explanations

    phrases related to environmental conservation, protection of ecosystems and wildlife, and human impact on biodiversity

    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
    1.3%
    575
    +0.15
    0.8%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    670
    +0.23
    0.11
    527
    +0.15
    0.02
    548
    +0.10
    0.01
    Negative Logits
    Machine
    -0.67
    urat
    -0.61
    otten
    -0.59
    Interstitial
    -0.59
    iaz
    -0.59
    avier
    -0.57
    ALSE
    -0.55
    ahon
    -0.55
    iffe
    -0.55
     Panzer
    -0.55
    POSITIVE LOGITS
     habitat
    1.02
     ecology
    0.99
     habitats
    0.96
     biologist
    0.93
     biodiversity
    0.91
     ecosystems
    0.87
     wildlife
    0.84
    conservancy
    0.84
     flora
    0.83
     conservation
    0.81
    Activations Density 17.644%

    No Known Activations