INDEX

Explanations

conclusive phrases that summarize or emphasize key points at the end of sections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ãģĭãĤı

-0.17

aju

-0.16

 Ø¬ÙĦ

-0.15

_skin

-0.15

ÑĥÐ¶

-0.15

.opend

-0.14

GBK

-0.14

.ud

-0.14

ëĭĺ

-0.14

ombine

-0.14

POSITIVE LOGITS

 Lastly

0.19

Lastly

0.18

finally

0.17

iversit

0.16

isci

0.16

ppo

0.16

Ð¾ÐºÑĢÐµÐ¼

0.15

 cooler

0.15

SCO

0.15

mente

0.15

Activations Density 0.022%

conclusive phrases that summarize or emphasize key points at the end of sections

No Comments

No Known Activations