INDEX

Explanations

punctuation marks indicating lists or items

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

haft

-0.15

Æ¡m

-0.15

Ð¾Ð±Ñī

-0.14

hta

-0.14

aison

-0.14

erala

-0.14

à¥įà¤°à¤¯

-0.14

bourne

-0.14

Ð»ÐµÑĤ

-0.14

ht

-0.13

POSITIVE LOGITS

ecs

0.15

ulty

0.15

Es

0.15

 Schmidt

0.13

dic

0.13

ease

0.13

 cans

0.13

Es

0.13

ajes

0.13

 ways

0.13

Activations Density 0.007%

punctuation marks indicating lists or items

No Comments

No Known Activations