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. 45072
    Prev
    Next
    INDEX
    Explanations

    the word "say" or "suggest" in statements within the text

    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
    +0.15
    0.8%
    274
    +0.15
    0.7%
    709
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    512
    +0.15
    0.09
    274
    +0.15
    0.10
    610
    +0.10
    0.06
    Negative Logits
     hemor
    -0.75
    pes
    -0.75
    xtap
    -0.69
    onut
    -0.68
     conflic
    -0.67
    itially
    -0.66
    Course
    -0.66
     millenn
    -0.65
    ¥ŀ
    -0.64
     fram
    -0.62
    POSITIVE LOGITS
     goodbye
    1.32
     hello
    0.98
    biz
    0.78
     sorry
    0.74
     Goodbye
    0.72
     aloud
    0.71
    ouse
    0.71
     farewell
    0.69
    dn
    0.68
    ieu
    0.65
    Activations Density 7.808%

    No Known Activations