INDEX

Explanations

references to "globe" or "global" concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

hips

-0.18

essler

-0.18

ë§ī

-0.18

 Territory

-0.17

eous

-0.17

rieve

-0.17

ORY

-0.16

ered

-0.15

auga

-0.14

gerald

-0.14

POSITIVE LOGITS

.glob

0.26

ular

0.26

ally

0.26

ule

0.22

 trot

0.22

-span

0.20

Ã¡lnÃŃ

0.20

ALLY

0.20

 Trot

0.19

ale

0.19

Activations Density 0.004%

references to "globe" or "global" concepts

No Comments

No Known Activations