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. 34154
    Prev
    Next
    INDEX
    Explanations

    references to academic disciplines, particularly in the humanities and social sciences, such as philosophy, history, and literature

    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: Mixed in middle - local

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.21
    1.0%
    288
    +0.15
    0.7%
    679
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.21
    0.12
    288
    +0.15
    0.22
    362
    +0.12
    0.10
    Negative Logits
    女
    -0.81
    ells
    -0.80
    esty
    -0.78
    cue
    -0.74
    ilee
    -0.71
    ookie
    -0.71
    rake
    -0.71
    bang
    -0.69
    bring
    -0.69
    elight
    -0.68
    POSITIVE LOGITS
    ophical
    1.32
    ophers
    0.96
    ophe
    0.87
    ophy
    0.87
    opher
    0.87
     philosopher
    0.83
    lectic
    0.81
    otle
    0.79
     philosophy
    0.78
     philosophers
    0.77
    Activations Density 10.137%

    No Known Activations