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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 15427
    Prev
    Next
    INDEX
    Explanations

    references to freshness, which could be related to food, news, or other contexts where something is recently made or obtained

    oai_token-act-pair · gpt-3.5-turbo

    references to freshness in food or products

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

    Negative Logits
    rael
    -0.70
    oried
    -0.69
    orset
    -0.64
    Ĥİ
    -0.61
     respecting
    -0.61
    orem
    -0.60
    ****
    -0.60
    asketball
    -0.60
    avery
    -0.60
    respect
    -0.60
    POSITIVE LOGITS
    ness
    1.19
    lish
    0.98
    faced
    0.89
    lings
    0.88
    water
    0.87
    foundland
    0.81
    lins
    0.80
    waters
    0.79
     fresh
    0.78
    bie
    0.76
    Activations Density 0.026%

    No Known Activations