© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 20201
    Prev
    Next
    INDEX
    Explanations

    prepositional phrases starting with 'in'

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    123
    +0.13
    0.6%
    640
    +0.12
    0.6%
    616
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.13
    0.06
    29
    +0.12
    0.06
    22
    +0.11
    0.05
    Negative Logits
    ATHER
    -0.60
     combatants
    -0.59
    weights
    -0.58
    76561
    -0.55
     embell
    -0.55
    tre
    -0.55
     awa
    -0.54
     gunmen
    -0.54
    ¶æ
    -0.54
    hur
    -0.53
    POSITIVE LOGITS
     spite
    1.41
     lieu
    1.31
     addition
    1.21
     accordance
    1.20
     retrospect
    1.14
    clus
    1.14
     essence
    1.13
    escap
    1.10
     conjunction
    1.09
    clusions
    1.09
    Activations Density 0.122%

    No Known Activations