INDEX

Explanations

references to intensive training or instructional programs

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

ona

-0.17

anko

-0.17

 Doctor

-0.15

quam

-0.15

onas

-0.14

CTOR

-0.14

ÄĽtÃŃ

-0.14

Ã¸r

-0.14

 WaitForSeconds

-0.14

mites

-0.14

POSITIVE LOGITS

elere

0.17

estone

0.17

adem

0.16

rios

0.15

ovu

0.14

ighter

0.14

 priv

0.14

 Else

0.13

aken

0.13

COD

0.13

Activations Density 0.003%

references to intensive training or instructional programs

No Comments

No Known Activations