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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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 Local
    5. 6-RES_SCL-AJT
    6. 40131
    Prev
    Next
    INDEX
    Explanations

    mentions of political figures, actions, and controversies

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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: Mostly-local line at bottom - local

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.20
    2.6%
    756
    +0.13
    1.7%
    459
    +0.13
    1.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.20
    -0.28
    459
    +0.13
    0.43
    87
    +0.13
    -0.36
    Negative Logits
    è¦ļéĨĴ
    -0.26
    cture
    -0.25
    lander
    -0.21
    ¶
    -0.20
    ©¶æ
    -0.20
    cedented
    -0.19
    acus
    -0.19
    opia
    -0.19
    podcast
    -0.19
    arent
    -0.18
    POSITIVE LOGITS
    anwhile
    0.23
     convertible
    0.20
     intest
    0.20
     replaced
    0.20
     substituted
    0.20
     ming
    0.19
     overhead
    0.19
     interchange
    0.19
     displ
    0.19
     entit
    0.19
    Activations Density 47.247%

    No Known Activations