© 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. 2-RES-JB
    11. 19408
    Prev
    Next
    INDEX
    Explanations

    passive voice constructions with verbs related to negative situations like neglect, exclusion, and inappropriate behavior

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to states of being or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    urry
    -0.64
    enture
    -0.63
     Moines
    -0.62
     Moons
    -0.61
    atl
    -0.61
     Lump
    -0.61
     Ago
    -0.60
     Yon
    -0.60
    ESE
    -0.59
     Fail
    -0.58
    POSITIVE LOGITS
    rael
    0.93
    senal
    0.92
     indistinguishable
    0.80
     nevertheless
    0.79
     treated
    0.77
     essentially
    0.75
     portrayed
    0.75
     not
    0.75
     shown
    0.74
     deemed
    0.73
    Activations Density 1.029%

    No Known Activations