INDEX

Explanations

references to specific contexts or scenarios where actions or events occur

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ã¡ce

-0.16

 nues

-0.16

.epam

-0.16

 ØµØ§Ø¯

-0.16

ernote

-0.15

Ð·Ñĭ

-0.15

legalArgumentException

-0.15

bote

-0.15

lod

-0.14

iales

-0.14

POSITIVE LOGITS

appropri

0.18

 appropriate

0.16

 unexpected

0.15

 otherwise

0.15

appropriate

0.15

ctors

0.14

otherwise

0.14

 most

0.14

 Sole

0.14

Activations Density 0.164%

references to specific contexts or scenarios where actions or events occur

No Comments

No Known Activations