© 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. 35122
    Prev
    Next
    INDEX
    Explanations

    phrases or words related to extraordinary or unusual events or situations

    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
    +0.21
    1.0%
    271
    +0.12
    0.6%
    722
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.21
    0.05
    622
    +0.12
    0.06
    669
    +0.12
    0.08
    Negative Logits
    Ķ
    -0.83
    hm
    -0.77
    rio
    -0.73
    Ö¼
    -0.72
    worn
    -0.72
    EEE
    -0.71
    hn
    -0.71
    Ĺ
    -0.71
    elsen
    -0.70
    gets
    -0.69
    POSITIVE LOGITS
    ordinary
    1.04
     rendition
    0.97
     circumstances
    0.95
    ities
    0.91
     circumstance
    0.88
    ional
    0.84
     occurrences
    0.82
     lengths
    0.79
     heroism
    0.78
    perty
    0.78
    Activations Density 10.280%

    No Known Activations