© 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
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 13648
    Prev
    Next
    INDEX
    Explanations

    phrases related to self-reflection and introspection

    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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    643
    +0.02
    0.5%
    676
    +0.02
    0.5%
    462
    +0.02
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    643
    +0.02
    0.09
    661
    +0.02
    0.09
    212
    +0.02
    0.10
    Negative Logits
     unlaw
    -0.14
     Spirits
    -0.13
     mushrooms
    -0.13
     targ
    -0.13
     conson
    -0.13
     tensions
    -0.13
     Presence
    -0.13
     tablets
    -0.13
     Galile
    -0.12
     mischief
    -0.12
    POSITIVE LOGITS
    ateurs
    0.21
    ateur
    0.21
    sterdam
    0.19
    estro
    0.19
    mediately
    0.18
    ethyst
    0.18
    nesty
    0.18
    entimes
    0.18
    ple
    0.17
    oru
    0.16
    Activations Density 7.242%

    No Known Activations