Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 625
    Prev
    Next
    INDEX
    Explanations

    critiques or discussions on various topics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    640
    +0.35
    1.7%
    64
    +0.13
    0.6%
    649
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.35
    0.29
    94
    +0.13
    0.20
    64
    +0.11
    -0.07
    Negative Logits
     footing
    -0.66
     Trog
    -0.65
     extrad
    -0.61
     equip
    -0.60
     communications
    -0.60
     charm
    -0.59
     Reincarnated
    -0.59
     hust
    -0.59
     blinded
    -0.59
    arde
    -0.58
    POSITIVE LOGITS
    ³³³³
    1.28
    ³³³³³³³³³³³³³³³³
    1.26
    ³³³³³³³³
    1.25
    ³³³
    1.24
    Anyway
    1.02
    Article
    1.01
    ³³
    0.99
    Advertisement
    0.96
    posted
    0.96
    Posted
    0.94
    Activations Density 1.318%

    No Known Activations