INDEX

Explanations

phrases related to incidents or experiences of assault and false accusations in a social setting

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tricot

-0.70

 nutella

-0.66

 matel

-0.64

 swarovski

-0.61

 frambo

-0.59

ihnachten

-0.58

 velours

-0.58

 mascarpone

-0.54

 ruine

-0.53

 sainte

-0.53

POSITIVE LOGITS

 incorrectly

0.86

 incorrect

0.83

<bos>

0.81

 erroneously

0.76

 mistakenly

0.72

 wrongly

0.68

 corrected

0.68

incorrect

0.67

 missp

0.66

 Incorrect

0.66

Activations Density 0.280%

phrases related to incidents or experiences of assault and false accusations in a social setting

No Comments

No Known Activations