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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 5098
    Prev
    Next
    INDEX
    Explanations

    future plans or intentions expressed in a decisive manner

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    393
    +0.27
    0.9%
    339
    +0.18
    0.6%
    608
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.27
    0.08
    49
    +0.18
    0.06
    608
    +0.15
    0.07
    Negative Logits
    cius
    -1.03
    VIDEO
    -0.90
    illin
    -0.89
    rising
    -0.88
    Reporting
    -0.87
    tymology
    -0.86
    Lear
    -0.85
    Reviewed
    -0.85
    usage
    -0.84
    mere
    -0.84
    POSITIVE LOGITS
     be
    1.51
     make
    1.27
     give
    1.22
     stick
    1.19
     come
    1.17
     get
    1.16
     lose
    1.14
     rely
    1.14
     see
    1.14
     earn
    1.14
    Activations Density 1.306%

    No Known Activations