INDEX

Explanations

programming-related terms and error handling patterns

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

.jquery

-0.14

Ã¢h

-0.14

ÑĢÐ°Ð½Ð¸ÑĨ

-0.14

GetInstance

-0.14

ëĮĢë¡ľ

-0.13

chal

-0.13

acus

-0.13

ÐłÐĺ

-0.13

lse

-0.13

-BEGIN

-0.13

POSITIVE LOGITS

apa

0.18

 conce

0.16

vere

0.16

zem

0.15

ateurs

0.15

err

0.15

.NoError

0.15

Zem

0.15

 Nile

0.15

(err

0.14

Activations Density 0.009%

programming-related terms and error handling patterns

No Comments

No Known Activations

programming-related terms and error handling patterns

No Comments

No Known Activations