INDEX

Explanations

quantitative data and statistics related to quantities and measurements

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

azy

-0.15

ari

-0.14

ellas

-0.14

³

-0.13

ussian

-0.13

viders

-0.13

Ĩµ

-0.13

HAVE

-0.13

 Westbrook

-0.13

 aspect

-0.13

POSITIVE LOGITS

 also

0.21

 again

0.19

also

0.18

ä¹Łä¸į

0.17

again

0.17

 teÅ¼

0.15

 Ø£ÙĬØ¶

0.15

èĭĹ

0.15

ionic

0.15

 ALSO

0.15

Activations Density 0.099%

quantitative data and statistics related to quantities and measurements

No Comments

No Known Activations