INDEX

Explanations

terms related to health and illness

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

âĢ¦↵↵↵

-0.14

iÐ²

-0.14

 Â·

-0.13

,âĢ¦

-0.12

 pioneer

-0.12

 Alabama

-0.12

_slices

-0.11

ecies

-0.11

Tow

-0.11

 ####

-0.11

POSITIVE LOGITS

raud

0.14

ëĿ¼ëıĦ

0.12

æ²Ł

0.12

ullan

0.12

jar

0.12

 Ð¿Ð¾Ð²

0.12

zelf

0.12

ãģ¯ãģļ

0.11

Fld

0.11

umba

0.11

Activations Density 0.140%

terms related to health and illness

No Comments

No Known Activations