© 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. 24082
    Prev
    Next
    INDEX
    Explanations

    words related to physical violence or destruction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the words "smashed" and "smashing" related to destruction or impact

    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
    ghan
    -0.71
    vernment
    -0.71
    anamo
    -0.68
    occup
    -0.67
    issue
    -0.66
    liction
    -0.63
    annis
    -0.62
    abet
    -0.60
    agonist
    -0.60
     volunteering
    -0.60
    POSITIVE LOGITS
     smashed
    1.05
     smash
    0.88
     smashing
    0.80
     shatter
    0.77
     shattered
    0.76
    ulence
    0.74
    busters
    0.72
    adoes
    0.72
     stomp
    0.71
     shards
    0.69
    Activations Density 0.013%

    No Known Activations