© 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. 26558
    Prev
    Next
    INDEX
    Explanations

    references to "Buffalo" and related entities in the text

    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
    269
    +0.13
    0.6%
    288
    +0.12
    0.5%
    608
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.13
    0.01
    269
    +0.12
    0.01
    117
    +0.11
    0.01
    Negative Logits
    ially
    -0.90
    egal
    -0.81
    arial
    -0.77
    olation
    -0.73
    enemy
    -0.73
    olate
    -0.73
    aft
    -0.72
    inen
    -0.71
    olated
    -0.71
    Mach
    -0.70
    POSITIVE LOGITS
     Trace
    1.05
     Sabres
    1.05
     Niagara
    1.04
     Bills
    1.02
     Bay
    0.78
     Knife
    0.77
     Springfield
    0.76
    town
    0.75
     Grove
    0.72
     Wild
    0.71
    Activations Density 0.005%

    No Known Activations