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 3072
    5. 8-RES_FS3072-JB
    6. 402
    Prev
    Next
    INDEX
    Explanations

    repeated occurrences of the word "flimsy"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    195
    +0.12
    0.6%
    314
    +0.12
    0.5%
    103
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.12
    0.04
    232
    +0.12
    0.04
    331
    +0.11
    0.04
    Negative Logits
    RAL
    -0.79
    selves
    -0.67
    TAIN
    -0.66
    HCR
    -0.65
    terday
    -0.65
     dummy
    -0.60
    OLOGY
    -0.59
     thumbs
    -0.59
     Territory
    -0.58
    eers
    -0.58
    POSITIVE LOGITS
    ickr
    1.24
    oyd
    1.21
    uffy
    1.19
    orescent
    1.17
    irting
    1.13
    oppy
    1.11
    orescence
    1.09
    icker
    1.08
    ushing
    1.06
    iers
    1.05
    Activations Density 0.442%

    No Known Activations