INDEX

Explanations

phrases indicating uncertainty or personal opinion

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

 poliuret

-0.90

 porcelaine

-0.86

zó

-0.83

 velours

-0.74

 republi

-0.73

 Præ

-0.72

 canne

-0.71

 vermel

-0.71

 majest

-0.70

 talle

-0.70

POSITIVE LOGITS

 there

0.62

 seems

0.59

 ExecuteAsync

0.58

animity

0.57

 that

0.54

 SOMEONE

0.51

ürze

0.51

 trône

0.50

 uitgenodigd

0.50

 suckers

0.50

Activations Density 0.086%

phrases indicating uncertainty or personal opinion

No Comments

No Known Activations