INDEX

Explanations

the word "originally" and its variations, indicating a focus on the origins or initial intentions behind subjects

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

avier

-0.18

al

-0.16

ackson

-0.16

ayne

-0.15

ifer

-0.15

amines

-0.15

ere

-0.15

 Williamson

-0.14

 William

-0.14

 Miller

-0.14

POSITIVE LOGITS

ãĥ¼ãĤ¹ãĥĪ

0.16

nings

0.16

ural

0.15

istrat

0.15

ochen

0.15

inals

0.15

_trampoline

0.14

yssey

0.14

bine

0.14

ĵåĲį

0.14

Activations Density 0.024%

the word "originally" and its variations, indicating a focus on the origins or initial intentions behind subjects

No Comments

No Known Activations