© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5334
    Prev
    Next
    INDEX
    Explanations

    verbs related to various types of actions and changes in states

    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
    494
    +0.14
    0.5%
    113
    +0.13
    0.5%
    303
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.14
    0.12
    665
    +0.13
    0.13
    282
    +0.13
    0.11
    Negative Logits
    dt
    -0.93
    eka
    -0.86
    tes
    -0.83
    heim
    -0.79
    ijk
    -0.78
    say
    -0.78
     laun
    -0.78
    racuse
    -0.77
    nton
    -0.76
    pel
    -0.76
    POSITIVE LOGITS
    ependent
    0.87
    edge
    0.85
    irection
    0.84
    Edge
    0.84
     Towards
    0.81
     Sparrow
    0.81
     ducks
    0.81
     upwards
    0.79
    itionally
    0.79
     poisoned
    0.79
    Activations Density 2.798%

    No Known Activations