© 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. 4085
    Prev
    Next
    INDEX
    Explanations

    information related to fundamental needs or concepts

    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
    +0.17
    0.7%
    99
    +0.14
    0.6%
    616
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.17
    0.24
    255
    +0.14
    0.09
    160
    +0.11
    0.17
    Negative Logits
    igham
    -0.84
    isSpecialOrderable
    -0.79
    romeda
    -0.74
     Tycoon
    -0.74
    crow
    -0.69
    uthor
    -0.69
     Rodrigo
    -0.68
    oping
    -0.67
    IDA
    -0.67
     Allaah
    -0.66
    POSITIVE LOGITS
     necessities
    1.18
     tenets
    1.11
    basic
    0.98
     arithmetic
    0.97
    lly
    0.94
     principles
    0.90
     outline
    0.89
    tenance
    0.86
     premise
    0.85
    gradient
    0.84
    Activations Density 12.541%

    No Known Activations