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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 7169
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Qu" followed by a digit (potentially part of the word Quilt)

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of 'Quorum.'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    HAEL
    -0.76
    AGES
    -0.75
     buds
    -0.69
     deed
    -0.67
    ãĤ¹ãĥĪ
    -0.64
     controller
    -0.63
    ãģĨ
    -0.63
     haze
    -0.62
     controllers
    -0.62
     warming
    -0.62
    POSITIVE LOGITS
    arantine
    1.19
    ests
    1.07
    ois
    1.05
    oted
    1.04
    atro
    0.99
    oting
    0.99
    icks
    0.98
    ipped
    0.96
    artz
    0.96
    isition
    0.96
    Activations Density 0.015%

    No Known Activations