© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 21467
    Prev
    Next
    INDEX
    Explanations

    phrases related to manual actions involving putting or attaching objects together

    oai_token-act-pair · gpt-3.5-turbo

    actions involving the verb "put" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SPONSORED
    -0.61
    Represent
    -0.61
    bred
    -0.59
    yan
    -0.58
    abiding
    -0.58
    ldom
    -0.58
     Heights
    -0.56
    progress
    -0.56
    journal
    -0.55
     perilous
    -0.55
    POSITIVE LOGITS
    tin
    0.91
     them
    0.87
     together
    0.87
     electrodes
    0.86
     aside
    0.80
     stickers
    0.80
     it
    0.76
     magnets
    0.75
    ãĥĥãĥī
    0.73
    »Ĵ
    0.72
    Activations Density 0.140%

    No Known Activations