INDEX

Explanations

references to health-related topics and well-being issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÑĤÐ¾ÑĢÐ°

-0.15

uck

-0.14

 Kosten

-0.14

 Costa

-0.14

áº§n

-0.13

asc

-0.13

 indeed

-0.13

-Ð³Ð¾

-0.13

ayet

-0.13

 <!--[

-0.13

POSITIVE LOGITS

 Moder

0.19

Moder

0.17

imbus

0.16

iminal

0.16

ENDOR

0.15

tbl

0.15

hunt

0.15

Ð·Ð´

0.14

ÑĥÐ¼Ñĥ

0.14

esa

0.14

Activations Density 0.031%

references to health-related topics and well-being issues

No Comments

No Known Activations