© 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. 20260
    Prev
    Next
    INDEX
    Explanations

    phrases related to various methods of communication and interaction

    oai_token-act-pair · gpt-3.5-turbo

    terms related to financial transactions and contractual agreements

    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
    bernatorial
    -0.72
    nai
    -0.67
    ippery
    -0.63
    ropolis
    -0.62
    irlf
    -0.58
    ergy
    -0.56
    amina
    -0.56
     wanted
    -0.56
    hillary
    -0.54
     feared
    -0.54
    POSITIVE LOGITS
     alone
    1.11
     rather
    0.91
     channels
    0.89
    rather
    0.86
     mechanisms
    0.75
     or
    0.74
     techniques
    0.74
     prism
    0.72
     referral
    0.72
     instead
    0.70
    Activations Density 0.411%

    No Known Activations