© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5898
    Prev
    Next
    INDEX
    Explanations

    detailed descriptions of materials and the processes associated with them, such as thin layers, welding, and semiconductor elements

    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
    195
    +0.21
    0.7%
    571
    +0.14
    0.5%
    35
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.21
    0.06
    326
    +0.14
    0.06
    288
    +0.13
    0.07
    Negative Logits
    pheus
    -1.07
    FAULT
    -0.98
    Particip
    -0.96
    NL
    -0.94
    leon
    -0.92
     Volt
    -0.90
    endants
    -0.86
    BA
    -0.80
    ILLE
    -0.80
     UNESCO
    -0.80
    POSITIVE LOGITS
    ned
    1.70
    ning
    1.51
    ners
    1.27
    nesses
    1.22
    ened
    1.18
    ciating
    1.17
     coats
    1.17
    ening
    1.17
    med
    1.15
    lings
    1.14
    Activations Density 0.858%

    No Known Activations