Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 1877
    Prev
    Next
    INDEX
    Explanations

    references to named entities, particularly those with 'NB' as a prefix followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the abbreviation "NB" that indicate various organizations or contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æ©
    -0.91
    ttle
    -0.78
    TYPE
    -0.76
    fare
    -0.76
     Ãľ
    -0.73
    tails
    -0.70
    eve
    -0.68
    «ĺ
    -0.65
    worms
    -0.65
    venge
    -0.63
    POSITIVE LOGITS
    NB
    0.93
    wik
    0.89
    ionics
    0.88
    emonic
    0.85
    ionic
    0.78
    eworthy
    0.78
    nb
    0.75
     NB
    0.73
    etheless
    0.71
    iotics
    0.70
    Activations Density 0.016%

    No Known Activations