INDEX

Explanations

references to insider information and stories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

à¹īà¸²à¸ĩ

-0.16

uo

-0.16

avia

-0.16

tiv

-0.14

 Burnett

-0.14

xic

-0.14

iamo

-0.14

alia

-0.14

 arch

-0.13

 appetite

-0.13

POSITIVE LOGITS

 HOUR

0.17

ãĥĥãĤ·ãĥ¥

0.17

 Hour

0.15

/out

0.15

pent

0.15

most

0.15

 Barrel

0.15

addtogroup

0.14

Hour

0.14

halb

0.14

Activations Density 0.009%

references to insider information and stories

No Comments

No Known Activations