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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1286
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial news and updates

    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
    70
    +0.11
    0.5%
    103
    +0.11
    0.5%
    696
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.11
    0.09
    131
    +0.11
    0.06
    696
    +0.10
    0.08
    Negative Logits
    paying
    -0.72
    ritional
    -0.69
    ileaks
    -0.68
     incentives
    -0.67
     trickle
    -0.67
    artifacts
    -0.66
    Progress
    -0.66
     timely
    -0.65
    ettings
    -0.64
    testing
    -0.64
    POSITIVE LOGITS
     slang
    1.54
     abbre
    1.52
     acronym
    1.48
     coined
    1.40
     shorthand
    1.36
     noun
    1.32
     adjective
    1.31
     suffix
    1.29
     pronunciation
    1.26
     initials
    1.25
    Activations Density 3.062%

    No Known Activations