INDEX

Explanations

instances of phrases indicating starting or initiating actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

.scalablytyped

-0.18

 Ø§ÙĪÙĦÛĮÙĨ

-0.18

 Starting

-0.17

åħ¥åı£

-0.17

Starting

-0.17

(first

-0.16

ç¬¬ä¸Ģ

-0.15

 starting

-0.15

 Entry

-0.15

 first

-0.14

POSITIVE LOGITS

 begin

0.52

begin

0.47

.begin

0.43

 Begin

0.41

 begins

0.40

Begin

0.40

 began

0.39

 BEGIN

0.37

BEGIN

0.35

 begun

0.35

Activations Density 0.019%

instances of phrases indicating starting or initiating actions

No Comments

No Known Activations