Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 5009
    Prev
    Next
    INDEX
    Explanations

    people's names, particularly focusing on the name "Adriana."

    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
    238
    +0.19
    0.6%
    332
    +0.15
    0.4%
    373
    +0.14
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.19
    0.04
    639
    +0.15
    0.04
    23
    +0.14
    0.04
    Negative Logits
    é¾
    -1.30
     Brunswick
    -1.12
     Pengu
    -1.03
     Moroc
    -1.02
     Fiorina
    -1.02
     chips
    -1.02
    manship
    -1.02
     magnification
    -1.01
    rights
    -1.00
     convol
    -0.99
    POSITIVE LOGITS
    ving
    1.76
    zzle
    1.74
    quet
    1.56
    quez
    1.55
    ety
    1.50
    ude
    1.41
    Äĩ
    1.41
    erer
    1.40
    zona
    1.39
    etta
    1.39
    Activations Density 0.526%

    No Known Activations