© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 5536
    Prev
    Next
    INDEX
    Explanations

    information related to geopolitical events, news, and journalism in Syria

    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
    365
    +0.15
    0.6%
    303
    +0.13
    0.5%
    258
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    145
    +0.15
    0.01
    373
    +0.13
    -0.00
    303
    +0.12
    0.01
    Negative Logits
    appa
    -0.79
    tein
    -0.77
    ineries
    -0.75
     Rail
    -0.75
    ilts
    -0.73
    uggest
    -0.73
    xit
    -0.73
     Cart
    -0.72
     Laf
    -0.69
    icken
    -0.69
    POSITIVE LOGITS
    today
    0.89
    hist
    0.84
    reality
    0.83
    bread
    0.81
    bi
    0.80
    bound
    0.79
    Stretch
    0.76
    iHUD
    0.76
    Pause
    0.75
    Face
    0.74
    Activations Density 0.002%

    No Known Activations