© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 37320
    Prev
    Next
    INDEX
    Explanations

    references to benefits or advantages in various contexts

    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
    420
    +0.13
    0.6%
    35
    +0.11
    0.5%
    575
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.13
    0.01
    724
    +0.11
    0.01
    225
    +0.11
    0.01
    Negative Logits
    issors
    -0.93
    entanyl
    -0.78
    chin
    -0.76
    izoph
    -0.73
    mind
    -0.71
    ichen
    -0.71
    adr
    -0.68
    cci
    -0.66
    ashes
    -0.66
    TW
    -0.66
    POSITIVE LOGITS
     perks
    1.08
     bestowed
    1.02
     entitle
    0.99
     conferred
    0.98
     afforded
    0.92
     perk
    0.92
     amenities
    0.88
     granted
    0.82
     privileges
    0.81
     rewards
    0.81
    Activations Density 0.015%

    No Known Activations