INDEX

Explanations

phrases indicating emphasis or focus on specific subjects or topics

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

zon

-0.17

alom

-0.15

à¹Ħà¸§

-0.14

tha

-0.14

æĮģãģ¡

-0.14

ieux

-0.14

urd

-0.13

 ÙģÙĨÛĮ

-0.13

udent

-0.13

.ibm

-0.13

POSITIVE LOGITS

 creampie

0.15

GGLE

0.15

 Shack

0.15

adas

0.15

omi

0.14

Cha

0.14

shed

0.14

emet

0.13

ozÃŃ

0.13

Activations Density 0.032%

phrases indicating emphasis or focus on specific subjects or topics

No Comments

No Known Activations