INDEX

Explanations

positive feedback and expressions of appreciation from audiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

-0.17

anke

-0.16

ierz

-0.16

ãĥĥãĤ«ãĥ¼

-0.16

 warn

-0.16

azÄĥ

-0.15

ÑĨÑĮÐºÐ¸Ð¹

-0.14

adena

-0.14

kish

-0.14

Ã±a

-0.14

POSITIVE LOGITS

insky

0.15

 Hend

0.15

 lÃ½

0.14

raith

0.14

(;;

0.14

 feedback

0.14

 come

0.14

à¥įà¤Ĺà¤¤

0.14

 bringing

0.13

 tÃŃm

0.13

Activations Density 0.165%

positive feedback and expressions of appreciation from audiences

No Comments

No Known Activations