© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 13809
    Prev
    Next
    INDEX
    Explanations

    self-referential language and expressions of personal frustration or disbelief

    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
    +0.18
    0.8%
    5
    +0.12
    0.5%
    257
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    527
    +0.18
    0.11
    679
    +0.12
    0.26
    267
    +0.10
    0.16
    Negative Logits
    rontal
    -0.71
    issors
    -0.64
    riger
    -0.64
     mirror
    -0.57
     Mori
    -0.56
    eday
    -0.56
    ALSE
    -0.56
     depreciation
    -0.56
     enactment
    -0.55
     microscope
    -0.55
    POSITIVE LOGITS
    aven
    0.65
    iple
    0.64
    iet
    0.63
    erville
    0.63
    soever
    0.63
    itle
    0.63
    bert
    0.62
    hawk
    0.62
    CU
    0.61
    loo
    0.61
    Activations Density 12.302%

    No Known Activations