© 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. 4844
    Prev
    Next
    INDEX
    Explanations

    mentions of length, depth, and types of physical measurements in 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
    +0.23
    1.0%
    166
    +0.15
    0.6%
    195
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    725
    +0.23
    0.07
    30
    +0.15
    0.04
    557
    +0.14
    0.07
    Negative Logits
    eers
    -1.07
    eer
    -0.95
    eering
    -0.83
    oras
    -0.77
    dinand
    -0.77
    tical
    -0.77
    unin
    -0.75
     Jarrett
    -0.72
    ILLE
    -0.72
    leck
    -0.70
    POSITIVE LOGITS
     length
    1.08
    shed
    1.08
     lengths
    1.06
    iest
    0.88
    length
    0.87
     circumference
    0.86
    wise
    0.84
    Length
    0.82
    running
    0.81
    sword
    0.81
    Activations Density 7.279%

    No Known Activations