INDEX

Explanations

references related to scientific research citations and DOI links

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bern

-0.15

GINE

-0.14

ene

-0.14

amp

-0.14

inton

-0.13

ÑİÑĢ

-0.13

ÐµÑĢÐº

-0.13

Ã±os

-0.13

unde

-0.13

_transient

-0.13

POSITIVE LOGITS

/--

0.14

imest

0.14

hi

0.14

odic

0.14

ãĤĤãģ£ãģ¨

0.13

 Rubin

0.13

Beh

0.13

²

0.13

 Goods

0.13

Activations Density 0.006%

references related to scientific research citations and DOI links

No Comments

No Known Activations