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. 4334
    Prev
    Next
    INDEX
    Explanations

    references to young children

    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
    153
    +0.16
    0.5%
    615
    +0.15
    0.5%
    711
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    433
    +0.16
    0.06
    711
    +0.15
    0.04
    302
    +0.15
    0.06
    Negative Logits
    headers
    -1.12
    anwhile
    -1.05
     mathemat
    -1.01
     tremend
    -1.01
    nce
    -0.98
    utherford
    -0.97
     confir
    -0.97
    chwitz
    -0.96
     conduc
    -0.95
    ainer
    -0.95
    POSITIVE LOGITS
     boy
    1.24
     girl
    1.11
     daughter
    1.08
     children
    1.06
     boys
    1.06
     toddler
    1.05
     toddlers
    1.03
     son
    1.01
     girls
    1.00
     child
    0.98
    Activations Density 0.831%

    No Known Activations