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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 11738
    Prev
    Next
    INDEX
    Explanations

    reviews and ratings given to products

    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
    274
    +0.16
    0.7%
    195
    +0.12
    0.5%
    630
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.16
    0.02
    609
    +0.12
    0.02
    630
    +0.10
    0.02
    Negative Logits
    ilant
    -0.74
    forth
    -0.71
    heny
    -0.68
     Nou
    -0.67
    adra
    -0.66
    ilus
    -0.65
    vous
    -0.64
    alter
    -0.63
    nown
    -0.63
    ãĤ¸
    -0.63
    POSITIVE LOGITS
    Rated
    1.02
    âĺħâĺħ
    0.96
    Rating
    0.92
    rating
    0.90
     rated
    0.85
    Recommend
    0.79
    Reviewed
    0.79
     rating
    0.76
    hyde
    0.70
    payers
    0.70
    Activations Density 0.045%

    No Known Activations