INDEX

Explanations

references to tanks or tank-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

SSIP

-0.18

 Carpenter

-0.15

è¸ı

-0.15

èŀº

-0.14

ãģĵãĤĵãģ«ãģ¡ãģ¯

-0.14

aux

-0.13

ource

-0.13

 narrowly

-0.13

 carbon

-0.13

POSITIVE LOGITS

oa

0.18

oop

0.16

nowled

0.16

akis

0.15

ÙĨØ¯Ùĩ

0.15

holder

0.15

phins

0.14

agement

0.14

yc

0.14

anoi

0.14

Activations Density 0.007%

references to tanks or tank-related topics

No Comments

No Known Activations