INDEX

Explanations

references to physical locations and settings

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

Ú¯ÛĮ

-0.19

venes

-0.18

 launder

-0.16

ë¶Ħ

-0.15

olla

-0.15

½

-0.15

wise

-0.14

inel

-0.14

agua

-0.14

arius

-0.14

POSITIVE LOGITS

cimal

0.17

ampoline

0.16

artz

0.16

borg

0.15

 zlib

0.15

PMC

0.15

ç¯

0.15

 slate

0.15

indle

0.15

kan

0.14

Activations Density 0.343%

references to physical locations and settings

No Comments

No Known Activations