INDEX

Explanations

phrases and words emphasizing the term "literally."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

agger

-0.17

ÙĪØ§Ø±Ùĩ

-0.17

emos

-0.14

 Ð¿Ñĥ

-0.14

();++

-0.14

furt

-0.14

/slick

-0.13

isphere

-0.13

 qued

-0.13

 normals

-0.13

POSITIVE LOGITS

 literally

0.27

literal

0.20

 literal

0.19

 Liter

0.18

-minded

0.18

Liter

0.18

liter

0.17

 hundreds

0.17

 Literal

0.17

mente

0.16

Activations Density 0.015%

phrases and words emphasizing the term "literally."

No Comments

No Known Activations