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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 46058
    Prev
    Next
    INDEX
    Explanations

    text related to rules and regulations in various fields such as sports, entertainment, and politics

    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: e2e-recon cluster above bottome line touching local cluster - local

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.27
    1.6%
    640
    +0.14
    0.9%
    679
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.27
    0.38
    640
    +0.14
    0.38
    6
    +0.09
    0.30
    Negative Logits
     unden
    -0.63
     challeng
    -0.58
     charact
    -0.56
     helicop
    -0.55
    onite
    -0.53
     neighb
    -0.53
     acknow
    -0.53
     predec
    -0.53
     skelet
    -0.51
     advoc
    -0.50
    POSITIVE LOGITS
    toggle
    0.69
     Subtle
    0.68
     Transcript
    0.67
     Whether
    0.64
     Planes
    0.63
     Its
    0.61
     Often
    0.59
     Located
    0.59
     Researchers
    0.58
     Contrary
    0.58
    Activations Density 15.603%

    No Known Activations