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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 12-RES-JB
    6. 186
    Prev
    Next
    INDEX
    Explanations

    instances of the infinitive form of verbs, particularly "to" followed by verb phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.12.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    12
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.52
    2.3%
    1334
    +0.10
    0.5%
    1967
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1415
    +0.52
    0.03
    1334
    +0.10
    0.03
    1425
    +0.10
    0.02
    Negative Logits
     intersper
    -1.34
    <bos>
    -1.01
     rouse
    -0.88
     overcrow
    -0.87
     endow
    -0.76
     depic
    -0.72
     banish
    -0.70
     gratify
    -0.69
     disambigu
    -0.69
     koz
    -0.68
    POSITIVE LOGITS
     pymongo
    0.84
     smtplib
    0.78
     heapq
    0.73
     pymysql
    0.73
     demas
    0.71
     ausp
    0.68
    بسم
    0.67
     barbacoa
    0.67
    SneakyThrows
    0.66
    ==""){
    0.65
    Activations Density 0.083%

    No Known Activations