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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 14567
    Prev
    Next
    INDEX
    Explanations

    words related to extreme or intense emotions or descriptions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "freak" and its variations along with expressions of intensity or emotional response

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adr
    -0.88
    eger
    -0.79
    ournal
    -0.73
    akespeare
    -0.68
    fman
    -0.68
    İĭ
    -0.67
     conduc
    -0.67
    dated
    -0.66
    cript
    -0.66
    rons
    -0.66
    POSITIVE LOGITS
    ishly
    1.27
    ously
    0.87
    onom
    0.86
     freak
    0.81
    wart
    0.79
    fuck
    0.79
    bum
    0.76
    ulously
    0.74
    istically
    0.73
    iness
    0.73
    Activations Density 0.008%

    No Known Activations