INDEX

Explanations

programming constructs and error handling in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

try

-0.20

_try

-0.18

Sharper

-0.16

Try

-0.16

try

-0.16

Try

-0.16

éŀ

-0.15

try

-0.14

è¯ķ

-0.14

 Beast

-0.14

POSITIVE LOGITS

ock

0.16

orman

0.15

ennifer

0.15

elman

0.14

stÅĻed

0.14

phis

0.14

¦

0.14

ëĿ¼ìĿ´

0.14

adge

0.14

 norm

0.14

Activations Density 0.018%

programming constructs and error handling in code

No Comments

No Known Activations