INDEX

Explanations

references to parent-child relationships within a coding context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

EDA

-0.17

ikel

-0.17

éĩ

-0.15

Ã¼h

-0.15

eda

-0.15

Î¼Î¿Ïħ

-0.15

 Ekon

-0.15

uzzer

-0.14

linkplain

-0.14

 volta

-0.14

POSITIVE LOGITS

Opened

0.16

dish

0.16

 opened

0.16

tit

0.15

gard

0.15

opened

0.14

Lon

0.14

ÌĤ

0.14

idity

0.14

nowledge

0.14

Activations Density 0.024%

references to parent-child relationships within a coding context

No Comments

No Known Activations