INDEX

Explanations

phrases indicating consistency or continuity throughout various experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

buz

-0.15

MBER

-0.15

 Disc

-0.15

/feed

-0.14

.selector

-0.14

 Select

-0.14

bai

-0.14

ewise

-0.14

PURE

-0.14

les

-0.13

POSITIVE LOGITS

 throughout

0.22

0.20

 suá»ĳt

0.20

 Throughout

0.19

Throughout

0.18

0.15

iah

0.15

áº¥u

0.14

ë°±

0.14

Activations Density 0.081%

phrases indicating consistency or continuity throughout various experiences

No Comments

No Known Activations