INDEX

Explanations

mentions of food-related terms like recipes and specific dishes

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>

-2.10

ⓧ

-0.77

<?

-0.71

-0.69

 effectually

-0.63

 guma

-0.59

 affatto

-0.57

/**

-0.56

grazie

-0.53

harmed

-0.53

POSITIVE LOGITS

Ins

1.40

Ins

1.31

INS

1.16

ins

1.13

ins

1.04

 Insulation

0.99

INS

0.99

 insulator

0.96

insu

0.95

 insulation

0.95

Activations Density 0.191%

mentions of food-related terms like recipes and specific dishes

No Comments

No Known Activations