INDEX

Explanations

numerical values related to measurements or quantities

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

reh

-0.15

ÅĻad

-0.15

hou

-0.14

Sphere

-0.14

 Ð²Ð¾ÑģÐ¿

-0.14

 repealed

-0.14

amba

-0.14

ertz

-0.14

dz

-0.13

POSITIVE LOGITS

ayer

0.15

 Bright

0.15

.scalablytyped

0.15

å¹¹ç·ļ

0.14

ram

0.14

ayers

0.14

Ð´ÑĢÐµÑģ

0.14

vero

0.14

ste

0.14

Bright

0.14

Activations Density 0.270%

numerical values related to measurements or quantities

No Comments

No Known Activations