INDEX

Explanations

warnings and cautionary advice related to tips, instructions, or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orz

-0.15

Äįan

-0.15

ecycle

-0.14

Ð¾Ð¹Ð½Ð¾

-0.14

 @"↵

-0.14

thinkable

-0.14

ATEGORIES

-0.14

imore

-0.14

ego

-0.14

aso

-0.13

POSITIVE LOGITS

 remember

0.49

remember

0.42

 Remember

0.38

Remember

0.38

 bear

0.38

be

0.38

don

0.38

 keep

0.36

 make

0.35

make

0.34

Activations Density 0.436%

warnings and cautionary advice related to tips, instructions, or actions

No Comments

No Known Activations