INDEX

Explanations

phrases that indicate empirical knowledge and traditional practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

mamak

-0.14

 Cotton

-0.14

 ØŃØ§Ø¶Ø±

-0.14

 xong

-0.14

("-

-0.14

ëĬ¥

-0.13

ìĩ¼

-0.13

STRU

-0.13

 Exercise

-0.13

 BindingFlags

-0.13

POSITIVE LOGITS

 already

0.24

already

0.22

Already

0.21

 Already

0.20

 knew

0.18

isd

0.17

 longtime

0.16

 earlier

0.16

_already

0.16

_ALREADY

0.15

Activations Density 0.164%

phrases that indicate empirical knowledge and traditional practices

No Comments

No Known Activations