INDEX

Explanations

elements related to progression and change in various contexts

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 annual

-0.23

 Annual

-0.20

annual

-0.18

Annual

-0.17

 weekly

-0.15

 yearly

-0.15

weekly

-0.15

 daily

-0.14

ÚĨÛĮ

-0.14

 monthly

-0.14

POSITIVE LOGITS

 each

1.02

each

0.94

Each

0.83

 Each

0.83

 EACH

0.79

 ÐºÐ°Ð¶Ð´

0.70

 cada

0.69

æ¯ı

0.67

 chaque

0.67

.each

0.65

Activations Density 0.393%

elements related to progression and change in various contexts

No Comments

No Known Activations