INDEX

Explanations

terms and phrases related to specialized expertise and professional designation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

isto

-0.15

frey

-0.15

setLayout

-0.15

 equally

-0.15

panic

-0.14

fulness

-0.14

nings

-0.14

que

-0.14

aler

-0.14

soever

-0.14

POSITIVE LOGITS

zed

0.19

à¹Ģà¸īà¸ŀà¸²à¸°

0.16

ritch

0.15

 jsonResponse

0.15

 Terrain

0.14

.Mutex

0.14

Ordinal

0.14

 Shir

0.14

terrain

0.13

Activations Density 0.020%

terms and phrases related to specialized expertise and professional designation

No Comments

No Known Activations