INDEX

Explanations

time indicators within the context of health or situation updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

dera

-0.16

elmet

-0.16

 Specialty

-0.15

Ð»Ð°ÑĤÐ°

-0.15

elf

-0.15

ilan

-0.14

ÏĦÎ¿Îº

-0.14

à¹Īà¸ĩ

-0.14

older

-0.14

lectric

-0.14

POSITIVE LOGITS

/all

0.15

uar

0.14

esian

0.14

asket

0.14

Nic

0.14

ãĥ¼ãĥĭ

0.14

lfw

0.13

ummings

0.13

Activations Density 0.050%

time indicators within the context of health or situation updates

No Comments

No Known Activations