© 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 Local
    5. 6-RES_SCL-AJT
    6. 45751
    Prev
    Next
    INDEX
    Explanations

    political content related to a specific individual and their actions

    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
    +0.27
    2.2%
    64
    +0.15
    1.2%
    756
    +0.12
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.27
    -0.12
    87
    +0.15
    -0.12
    526
    +0.12
    -0.08
    Negative Logits
     Originally
    -0.45
     Completed
    -0.40
    cohol
    -0.39
     Located
    -0.37
     Optical
    -0.37
     Previously
    -0.36
     Normally
    -0.35
     BMC
    -0.35
     Previous
    -0.35
     Click
    -0.35
    POSITIVE LOGITS
     scapego
    0.52
     damned
    0.51
     repud
    0.49
     goddamn
    0.47
     neocons
    0.47
    etheless
    0.47
     acquies
    0.46
     genuinely
    0.46
     prud
    0.46
     complicit
    0.46
    Activations Density 11.176%

    No Known Activations