INDEX

Explanations

the word "with" indicating partnerships or collaborations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

indre

-0.18

Ã¶r

-0.15

dojo

-0.15

phan

-0.15

 curs

-0.15

axon

-0.14

rx

-0.14

 quot

-0.14

IGO

-0.14

 nackt

-0.14

POSITIVE LOGITS

ansk

0.17

ABI

0.16

abi

0.16

otta

0.15

DTV

0.15

 blackout

0.14

eks

0.14

oldt

0.14

 reserv

0.14

Activations Density 0.145%

the word "with" indicating partnerships or collaborations

No Comments

No Known Activations