© 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. 14090
    Prev
    Next
    INDEX
    Explanations

    specific terms or words with a numerical value mentioned in the text

    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
    461
    +0.15
    0.5%
    603
    +0.13
    0.4%
    395
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    461
    +0.15
    0.05
    411
    +0.13
    0.05
    647
    +0.12
    0.04
    Negative Logits
    jri
    -1.08
    âĹ¼
    -1.01
    iland
    -1.00
    CI
    -0.96
    aukee
    -0.96
    erker
    -0.95
    GoldMagikarp
    -0.92
    ulhu
    -0.92
    _-
    -0.91
    gaard
    -0.91
    POSITIVE LOGITS
     itself
    1.45
    ultimate
    1.11
     oneself
    1.06
     yourself
    1.00
     himself
    1.00
    lessness
    1.00
    icide
    0.99
     '
    0.99
    marked
    0.98
    mark
    0.94
    Activations Density 0.698%

    No Known Activations