© 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. 37209
    Prev
    Next
    INDEX
    Explanations

    occurrences of the preposition "in" indicating 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
    481
    +0.35
    1.6%
    363
    +0.12
    0.6%
    372
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.35
    0.04
    363
    +0.12
    0.03
    526
    +0.11
    0.00
    Negative Logits
    äºĶ
    -0.60
     Minotaur
    -0.59
    ½
    -0.59
    define
    -0.58
    Definition
    -0.57
    Ľ
    -0.57
    ****
    -0.57
    UID
    -0.57
     Frey
    -0.56
    *****
    -0.53
    POSITIVE LOGITS
     workplaces
    0.81
     conferences
    0.81
    reads
    0.77
    Pub
    0.76
     classrooms
    0.76
     rooft
    0.76
     cafes
    0.75
     pubs
    0.73
     clinics
    0.71
    animate
    0.69
    Activations Density 0.177%

    No Known Activations