INDEX

Explanations

vocabulary related to financial evaluations and losses

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

########.

-0.71

parsedMessage

-0.70

 оригіналу

-0.70

 चीज़ों

-0.67

yntaxException

-0.66

 informée

-0.65

 ProtoMessage

-0.65

 мәкал

-0.64

hoeddwyd

-0.64

஦

-0.63

POSITIVE LOGITS

0.35

 over

0.31

 after

0.31

 further

0.29

 mountain

0.28

都

0.27

ra

0.27

 Frey

0.27

“

0.27

 continued

0.27

Activations Density 0.088%

vocabulary related to financial evaluations and losses

No Comments

No Known Activations