INDEX

Explanations

various forms of the word "inspire" and its related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ilde

-0.19

ierz

-0.17

stile

-0.17

liness

-0.16

ÏĦÎ±Î½

-0.15

Ð·Ð²

-0.15

advisor

-0.15

essa

-0.15

esser

-0.15

.habbo

-0.15

POSITIVE LOGITS

ally

0.16

zia

0.16

 others

0.16

 confidence

0.15

 hope

0.15

 quotes

0.15

awe

0.15

apl

0.15

others

0.14

 greatness

0.14

Activations Density 0.036%

various forms of the word "inspire" and its related concepts

No Comments

No Known Activations