© 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. 14108
    Prev
    Next
    INDEX
    Explanations

    verbs indicating actions or behavior

    oai_token-act-pair · gpt-3.5-turbo

    negative statements about actions or experiences

    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
    accompan
    -0.76
    unknown
    -0.74
    accompanied
    -0.66
    abre
    -0.66
     unparalleled
    -0.65
    ãĤ¼ãĤ¦ãĤ¹
    -0.65
    moil
    -0.64
    pmwiki
    -0.64
    avoid
    -0.64
    andan
    -0.63
    POSITIVE LOGITS
     anymore
    1.69
     anything
    1.36
     nor
    1.26
     any
    1.21
     anybody
    1.17
     anywhere
    1.09
     enough
    1.08
     ANY
    1.05
     slightest
    1.00
     anyone
    0.99
    Activations Density 0.240%

    No Known Activations