INDEX

Explanations

phrases indicating outcomes or actions that build towards improvement or change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ENSITY

-0.16

ints

-0.15

ilater

-0.15

ÙħÙĨ

-0.14

rah

-0.13

ecd

-0.13

iw

-0.13

Ã¼h

-0.13

@brief

-0.13

arih

-0.13

POSITIVE LOGITS

 increased

0.19

 further

0.17

 greater

0.17

 being

0.17

 eventual

0.16

 decreased

0.16

 conclusion

0.15

 interesting

0.15

 loss

0.14

 predictable

0.14

Activations Density 0.096%

phrases indicating outcomes or actions that build towards improvement or change

No Comments

No Known Activations