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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 13910
    Prev
    Next
    INDEX
    Explanations

    phrases related to attending events or meetings

    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
    •Layer 6 UMAP region: Mixed in middle - e2e-recon

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    445
    +0.13
    0.5%
    246
    +0.12
    0.4%
    504
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.13
    0.04
    469
    +0.12
    0.03
    555
    +0.12
    0.03
    Negative Logits
    arin
    -0.85
    theless
    -0.81
    Tro
    -0.79
    slice
    -0.79
    obil
    -0.78
    cold
    -0.78
    ortium
    -0.77
    pmwiki
    -0.76
    vc
    -0.75
    worn
    -0.73
    POSITIVE LOGITS
     RTX
    1.04
     meetings
    1.02
     attend
    0.95
     Attend
    0.94
    eret
    0.94
     conventions
    0.92
     hearings
    0.91
     conferences
    0.90
    URN
    0.86
    ees
    0.85
    Activations Density 0.778%

    No Known Activations