© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3639
    Prev
    Next
    INDEX
    Explanations

    words related to negative experiences or emotions, such as frustration, boredom, and annoyance

    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
    326
    +0.19
    0.7%
    271
    +0.18
    0.6%
    453
    +0.16
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.19
    0.09
    151
    +0.18
    0.10
    453
    +0.16
    0.07
    Negative Logits
    orah
    -1.01
    igham
    -0.97
    avia
    -0.96
    flags
    -0.95
    apor
    -0.94
    heng
    -0.93
    idelines
    -0.91
    ortium
    -0.90
    orthy
    -0.90
    leaf
    -0.90
    POSITIVE LOGITS
    ly
    1.35
     ted
    1.24
    ingly
    1.09
     repetition
    1.03
     depress
    0.94
     stare
    0.91
    ãĥ¼ãĥĨ
    0.91
    entimes
    0.90
     slog
    0.89
     repetitive
    0.88
    Activations Density 1.132%

    No Known Activations