© 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. 26530
    Prev
    Next
    INDEX
    Explanations

    terms related to job positions and work responsibilities

    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
    710
    +0.11
    0.5%
    42
    +0.10
    0.5%
    504
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.11
    0.04
    469
    +0.10
    0.02
    295
    +0.10
    0.02
    Negative Logits
     disapp
    -0.75
    ulic
    -0.66
    otine
    -0.62
     scram
    -0.61
     decre
    -0.61
     prud
    -0.60
     impulse
    -0.60
     landslide
    -0.59
     confidence
    -0.59
     avoidance
    -0.59
    POSITIVE LOGITS
    Behind
    0.84
     abroad
    0.80
     elsewhere
    0.79
    Intern
    0.78
     alongside
    0.76
    DragonMagazine
    0.76
    Cover
    0.74
    hare
    0.74
    ãĥŀ
    0.71
    çīĪ
    0.70
    Activations Density 0.138%

    No Known Activations