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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 7524
    Prev
    Next
    INDEX
    Explanations

    adverbs expressing contrast or exception

    oai_token-act-pair · gpt-3.5-turbo

    the use of the word "albeit" or similar structures that introduce qualifying or contrasting statements

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    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
    393
    +0.22
    0.7%
    547
    +0.13
    0.4%
    706
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.22
    0.09
    208
    +0.13
    0.09
    236
    +0.13
    0.09
    Negative Logits
    pta
    -0.98
    ORPG
    -0.96
    vance
    -0.93
    enium
    -0.89
    士
    -0.89
    arthed
    -0.88
    ving
    -0.87
     Medals
    -0.86
     Europa
    -0.83
     Pigs
    -0.83
    POSITIVE LOGITS
    tons
    1.42
     nonetheless
    1.33
    chery
    1.31
     nevertheless
    1.21
     fortunately
    1.18
     uncom
    1.11
     ultimately
    1.11
    tered
    1.10
    kus
    1.08
     subtle
    1.07
    Activations Density 1.361%

    No Known Activations