INDEX

Explanations

phrases related to product specifications or features

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.78

-1.00

ⓧ

-0.85

<?

-0.79

/**

-0.77

displayquote

-0.69

/***

-0.67

/*++

-0.67

Будь

-0.66

 AssemblyCompany

-0.65

POSITIVE LOGITS

 wien

1.43

 bangkok

1.38

 bandung

1.35

 jaya

1.32

 ankara

1.31

 Minang

1.30

 disreg

1.29

 thut

1.26

olx

1.25

 dises

1.21

Activations Density 0.132%

phrases related to product specifications or features

No Comments

No Known Activations