INDEX

Explanations

experiences related to learning and teaching skills

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jedis

-0.55

Literatuur

-0.54

 hại

-0.54

twimg

-0.53

chester

-0.53

 समीक्षाएं

-0.52

rachtet

-0.51

yelidikan

-0.51

ouncements

-0.50

 result

-0.50

POSITIVE LOGITS

 useStyles

0.70

 gelernt

0.69

 techniques

0.66

 taught

0.65

parsedMessage

0.62

 Taught

0.61

 Umgang

0.61

perfectly

0.60

bootstrapcdn

0.60

taught

0.60

Activations Density 0.062%

experiences related to learning and teaching skills

No Comments

No Known Activations