© 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. 4149
    Prev
    Next
    INDEX
    Explanations

    technical terms related to computer networking and technology, such as "firewall", "subnet", and "NSX"

    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
    167
    +0.14
    0.6%
    103
    +0.13
    0.6%
    739
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.15
    302
    +0.13
    0.09
    103
    +0.12
    0.07
    Negative Logits
    glers
    -1.31
    swick
    -0.89
     âĨij
    -0.75
     Brus
    -0.75
    ters
    -0.73
    ppo
    -0.71
    GY
    -0.71
    icals
    -0.71
    terday
    -0.71
    gery
    -0.71
    POSITIVE LOGITS
    ributed
    1.49
    inguished
    1.32
    ribute
    1.29
    ribution
    1.27
    ribut
    1.23
    inct
    1.16
    rict
    1.13
    inctions
    1.12
    ortion
    1.05
    inction
    0.96
    Activations Density 4.859%

    No Known Activations