© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 1434
    Prev
    Next
    INDEX
    Explanations

    phrases related to education, government funding, and financial cuts

    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
    288
    +0.14
    0.5%
    636
    +0.12
    0.4%
    575
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.14
    0.17
    288
    +0.12
    0.19
    481
    +0.12
    0.18
    Negative Logits
    hov
    -0.93
     Greenberg
    -0.78
    soType
    -0.74
    Cat
    -0.72
    ilingual
    -0.72
     Vend
    -0.72
     Plenty
    -0.71
     Veterinary
    -0.71
     Cats
    -0.70
    chn
    -0.70
    POSITIVE LOGITS
     altogether
    1.43
     prematurely
    1.16
     entirely
    1.11
     inhib
    1.11
     barrier
    0.98
    til
    0.92
     outright
    0.92
     unnecessarily
    0.90
    avorite
    0.88
     abruptly
    0.87
    Activations Density 4.663%

    No Known Activations