INDEX

Explanations

phrases that indicate totality or completeness, particularly the word "all" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

-0.16

edition

-0.14

multipart

-0.14

LOGGER

-0.14

.tm

-0.14

 Null

-0.13

olvers

-0.13

irus

-0.13

leich

-0.13

celed

-0.13

POSITIVE LOGITS

 Hack

0.15

ibo

0.15

cales

0.15

BK

0.15

–↵↵

0.14

RIPT

0.14

 Podle

0.14

ude

0.14

ë¡ľìĦľ

0.14

Ø§ØŃ

0.14

Activations Density 0.085%

phrases that indicate totality or completeness, particularly the word "all" in various contexts

No Comments

No Known Activations