© 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. 14160
    Prev
    Next
    INDEX
    Explanations

    phrases related to fulfilling promises or obligations

    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
    139
    +0.10
    0.5%
    288
    +0.10
    0.5%
    540
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.05
    151
    +0.10
    0.05
    288
    +0.10
    0.06
    Negative Logits
    oxin
    -0.64
    arij
    -0.64
    perature
    -0.62
    nesty
    -0.60
    acea
    -0.60
    uyomi
    -0.59
     separating
    -0.58
     Acid
    -0.57
    apter
    -0.56
    maxwell
    -0.56
    POSITIVE LOGITS
     fulfil
    0.77
     fulfillment
    0.77
     fulf
    0.76
     fulfilled
    0.75
    ãĥĺ
    0.74
    kj
    0.72
    icles
    0.68
    OGR
    0.68
     roles
    0.67
     fulfill
    0.66
    Activations Density 0.637%

    No Known Activations