© 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 Local
    5. 6-RES_SCL-AJT
    6. 40672
    Prev
    Next
    INDEX
    Explanations

    technical details and specifications related to mechanical components and their performance

    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
    314
    +0.10
    0.7%
    640
    +0.09
    0.6%
    393
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.10
    0.18
    22
    +0.09
    0.19
    29
    +0.09
    0.19
    Negative Logits
    soever
    -0.47
    killed
    -0.43
     hous
    -0.43
    interrupted
    -0.42
     %%
    -0.40
     deported
    -0.40
     vulner
    -0.40
     unemploy
    -0.39
     raped
    -0.38
     lodged
    -0.38
    POSITIVE LOGITS
    clusions
    0.83
    escap
    0.80
     regards
    0.77
    clus
    0.74
    efficiency
    0.73
    ordinate
    0.72
     accordance
    0.72
     favor
    0.71
    effic
    0.69
     lieu
    0.69
    Activations Density 8.093%

    No Known Activations