© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 5861
    Prev
    Next
    INDEX
    Explanations

    quotes and dialogues in sentences or conversation format

    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
    68
    +0.14
    0.5%
    482
    +0.10
    0.4%
    247
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.14
    0.27
    154
    +0.10
    0.26
    352
    +0.10
    0.25
    Negative Logits
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.99
    anwhile
    -0.96
     tremend
    -0.93
    lda
    -0.92
     eleph
    -0.87
     foreign
    -0.82
     domestic
    -0.81
    ãĥīãĥ©ãĤ´ãĥ³
    -0.79
     mosqu
    -0.78
     Leban
    -0.77
    POSITIVE LOGITS
    º
    1.10
    ¹
    1.05
    £
    1.02
    Ĭ
    1.00
    į
    0.99
    ¡
    0.98
    ¿
    0.98
    thouse
    0.97
    ı
    0.96
    atures
    0.96
    Activations Density 1.542%

    No Known Activations