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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 45525
    Prev
    Next
    INDEX
    Explanations

    terms related to database migration processes and configurations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    480
    +0.46
    2.4%
    326
    +0.13
    0.7%
    640
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.46
    0.04
    176
    +0.13
    0.03
    6
    +0.12
    0.04
    Negative Logits
    conservancy
    -0.81
     reluct
    -0.71
     phenomen
    -0.58
     encount
    -0.57
     antidepress
    -0.56
     growing
    -0.56
     veter
    -0.56
     entreprene
    -0.56
    Ó
    -0.56
     undermin
    -0.55
    POSITIVE LOGITS
     =================================================================
    0.89
    endif
    0.84
     ********************************
    0.81
     ----------------------------------------------------------------
    0.78
     <-
    0.78
    ))))
    0.78
     Unsure
    0.76
     Conclusion
    0.74
     WRITE
    0.73
     Repeat
    0.71
    Activations Density 0.182%

    No Known Activations