INDEX

Explanations

terms related to the status of code execution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

↵

-0.16

ody

-0.16

ãĤ¤ãĤ¯

-0.16

eft

-0.15

_styles

-0.15

areth

-0.14

_aliases

-0.14

istrat

-0.14

to

-0.14

ager

-0.14

POSITIVE LOGITS

quo

0.49

ses

0.31

quo

0.28

(es

0.22

lights

0.22

sing

0.22

sed

0.21

phere

0.21

es

0.20

 epile

0.20

Activations Density 0.017%

terms related to the status of code execution

No Comments

No Known Activations