Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 33269
    Prev
    Next
    INDEX
    Explanations

    common keywords related to software development and programming languages

    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
    +0.34
    1.5%
    190
    +0.10
    0.4%
    692
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.34
    0.35
    739
    +0.10
    0.04
    644
    +0.10
    0.14
    Negative Logits
    attled
    -0.73
    ously
    -0.68
    20439
    -0.67
    hyde
    -0.67
    feld
    -0.66
    ivably
    -0.66
    equal
    -0.65
    Brit
    -0.65
    eday
    -0.64
    mson
    -0.63
    POSITIVE LOGITS
     Definitions
    1.14
     Overview
    1.12
    Overview
    1.10
     nutshell
    0.94
     Background
    0.94
     Introduction
    0.91
     Basics
    0.91
     Differences
    0.91
     Summary
    0.90
     basics
    0.86
    Activations Density 23.245%

    No Known Activations