INDEX

Explanations

references to research findings and results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

.TestTools

-0.14

ADDE

-0.14

 âĹĦ

-0.13

ÐµÑĤÐ¸

-0.13

atus

-0.13

ÃĹ</

-0.13

Ø®Ø§ÙĨÙĩ

-0.13

à¥ĩà¤¯

-0.13

ÐµÐ²Ð¸Ñĩ

-0.13

lec

-0.13

POSITIVE LOGITS

urations

0.17

edl

0.16

/response

0.16

hoot

0.15

ereco

0.14

finder

0.14

IMA

0.14

nable

0.13

sup

0.13

edly

0.13

Activations Density 0.037%

references to research findings and results

No Comments

No Known Activations