© 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. 34800
    Prev
    Next
    INDEX
    Explanations

    sections of code related to setting up and executing functions in a programming environment

    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
    +0.12
    0.6%
    288
    +0.11
    0.6%
    437
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.23
    575
    +0.11
    0.16
    326
    +0.10
    0.22
    Negative Logits
    abouts
    -0.79
    soever
    -0.76
     seiz
    -0.74
    theless
    -0.71
    humans
    -0.70
    lihood
    -0.68
    kos
    -0.67
    eatures
    -0.67
     resil
    -0.67
    ternity
    -0.67
    POSITIVE LOGITS
    Flags
    0.82
     Context
    0.81
    Info
    0.80
    Manager
    0.79
    Types
    0.79
    Config
    0.77
    Parameters
    0.77
    Instance
    0.77
    Type
    0.76
     Builder
    0.76
    Activations Density 9.355%

    No Known Activations