INDEX

Explanations

references to relationships and positions, particularly in professional contexts

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

.Guna

-0.16

.scalablytyped

-0.16

 Plantae

-0.15

ìĤ

-0.15

$MESS

-0.15

taboola

-0.14

enderror

-0.14

tzv

-0.14

.gnu

-0.14

okud

-0.14

POSITIVE LOGITS

0.18

0.17

—and

0.16

0.15

se

0.15

 both

0.15

 dual

0.15

↵

0.15

 initial

0.15

Activations Density 0.138%

references to relationships and positions, particularly in professional contexts

No Comments

No Known Activations