© 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
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 4884
    Prev
    Next
    INDEX
    Explanations

    numerical values or measurements in a technical context

    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
    739
    +0.07
    0.5%
    701
    +0.06
    0.5%
    670
    +0.05
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    739
    +0.07
    0.06
    71
    +0.06
    0.03
    640
    +0.05
    0.19
    Negative Logits
     Khe
    -0.44
     patri
    -0.40
     duly
    -0.37
    ikuman
    -0.36
     Beir
    -0.35
     estab
    -0.35
    ãĥ¼ãĥĨãĤ£
    -0.35
     WATCHED
    -0.35
    endants
    -0.34
     Zeit
    -0.34
    POSITIVE LOGITS
    ...]
    0.65
    edit
    0.51
    âĢ¦]
    0.50
    especially
    0.49
    !--
    0.48
    ËĪ
    0.48
    emphasis
    0.47
    Pg
    0.47
    rule
    0.47
    tymology
    0.44
    Activations Density 8.093%

    No Known Activations